From ce008d5b81216c406f79588315cf3b5b08c5ed48 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 27 Sep 2023 16:11:57 +0800 Subject: [PATCH 01/48] Add files via upload --- Hina_Mod_AICoverGen_colab.ipynb | 308 ++++++++++++++++++++++++++++++++ 1 file changed, 308 insertions(+) create mode 100644 Hina_Mod_AICoverGen_colab.ipynb diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb new file mode 100644 index 00000000..d559dd97 --- /dev/null +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -0,0 +1,308 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "39k2mOCNAh6J" + }, + "source": [ + "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SociallyIneptWeeb/AICoverGen/blob/main/AICoverGen_colab.ipynb)" + ] + }, + { + "cell_type": "markdown", + "source": [ + "# AICoverGen WebUI\n", + "\n", + "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." + ], + "metadata": { + "id": "YYVAKuNBc-X4" + } + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "id": "vC4gLMHI9xb3", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "a839f0a7-e240-4020-f80a-e54fe3d27051" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Done Cloning Repository\n" + ] + } + ], + "source": [ + "#@title Clone repository\n", + "from IPython.display import clear_output, Javascript\n", + "import codecs\n", + "cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", + "!git clone $cloneing\n", + "%cd AICoverGen\n", + "clear_output()\n", + "print(\"Done Cloning Repository\")" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "odzpJHpr_PaF", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "364aac42-6448-4752-bfff-3c32af3e49f4" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Finsihed running this cell, proceed to the next cell\n" + ] + } + ], + "source": [ + "#@title Install requirements\n", + "!pip install -q -r requirements.txt\n", + "clear_output()\n", + "print(\"Finished Installing Requirements\")\n", + "!sudo apt update\n", + "clear_output()\n", + "print(\"Finished Updating\")\n", + "!sudo apt install sox\n", + "clear_output()\n", + "print(\"Finsihed running this cell, proceed to the next cell\")" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "SLWpcJc0AHSZ", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "56b66089-0ae3-462a-c938-7e96e88044e5" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Finished Downloading Voice Separation Model and Hubert Base Model\n" + ] + } + ], + "source": [ + "#@title Download MDXNet Vocal Separation and Hubert Base Models\n", + "!python src/download_models.py\n", + "clear_output()\n", + "print(\"Finished Downloading Voice Separation Model and Hubert Base Model\")" + ] + }, + { + "cell_type": "code", + "source": [ + "#@title Run WebUI\n", + "runpy=codecs.decode('fep/jrohv.cl','rot_13')\n", + "!python $runpy --share" + ], + "metadata": { + "id": "NEglTq6Ya9d0", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "617f2b65-5e54-438c-d991-a170e2ce9b7c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "2023-09-27 07:07:48.631125: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", + "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", + "2023-09-27 07:07:49.538446: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", + "2023-09-27 07:07:50 | INFO | fairseq.tasks.text_to_speech | Please install tensorboardX: pip install tensorboardX\n", + "2023-09-27 07:07:51 | INFO | faiss.loader | Loading faiss with AVX2 support.\n", + "2023-09-27 07:07:51 | INFO | faiss.loader | Successfully loaded faiss with AVX2 support.\n", + "Running on local URL: http://127.0.0.1:7860\n", + "Running on public URL: https://513f29c74e8615004f.gradio.live\n", + "\n", + "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n", + "2023-09-27 07:08:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:08:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:08:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:08:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:09:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:09:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:09:28 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:09:28 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 38/38 [00:11<00:00, 3.17it/s]\n", + "100% 38/38 [00:10<00:00, 3.71it/s]\n", + "100% 38/38 [00:07<00:00, 5.09it/s]\n", + "100% 38/38 [00:08<00:00, 4.26it/s]\n", + "100% 20/20 [00:11<00:00, 1.74it/s]\n", + "100% 20/20 [00:13<00:00, 1.49it/s]\n", + "2023-09-27 07:13:21 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:13:21 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:13:21 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:13:55 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:13:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 38/38 [00:10<00:00, 3.66it/s]\n", + "100% 38/38 [00:12<00:00, 3.08it/s]\n", + "100% 38/38 [00:08<00:00, 4.62it/s]\n", + "100% 38/38 [00:08<00:00, 4.61it/s]\n", + "100% 20/20 [00:12<00:00, 1.59it/s]\n", + "100% 20/20 [00:11<00:00, 1.73it/s]\n", + "2023-09-27 07:18:32 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:18:32 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:18:32 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:19:05 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:19:05 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 48/48 [00:13<00:00, 3.58it/s]\n", + "100% 48/48 [00:14<00:00, 3.40it/s]\n", + "100% 48/48 [00:10<00:00, 4.80it/s]\n", + "100% 48/48 [00:11<00:00, 4.13it/s]\n", + "100% 24/24 [00:13<00:00, 1.72it/s]\n", + "100% 24/24 [00:13<00:00, 1.74it/s]\n", + "2023-09-27 07:22:40 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:22:40 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:22:40 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:23:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:23:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 48/48 [00:13<00:00, 3.63it/s]\n", + "100% 48/48 [00:13<00:00, 3.51it/s]\n", + "100% 48/48 [00:10<00:00, 4.65it/s]\n", + "100% 48/48 [00:11<00:00, 4.12it/s]\n", + "100% 24/24 [00:13<00:00, 1.76it/s]\n", + "100% 24/24 [00:13<00:00, 1.79it/s]\n", + "2023-09-27 07:32:19 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:32:19 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:32:19 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:32:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:32:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 46/46 [00:12<00:00, 3.59it/s]\n", + "100% 46/46 [00:13<00:00, 3.31it/s]\n", + "100% 46/46 [00:09<00:00, 4.76it/s]\n", + "100% 46/46 [00:11<00:00, 4.01it/s]\n", + "100% 24/24 [00:13<00:00, 1.71it/s]\n", + "100% 24/24 [00:14<00:00, 1.68it/s]\n", + "2023-09-27 07:35:42 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:35:42 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:35:42 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:36:07 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:36:07 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:36:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:36:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:38:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:38:45 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:39:13 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:39:14 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 52/52 [00:14<00:00, 3.57it/s]\n", + "100% 52/52 [00:14<00:00, 3.53it/s]\n", + "100% 52/52 [00:12<00:00, 4.28it/s]\n", + "100% 52/52 [00:11<00:00, 4.68it/s]\n", + "100% 26/26 [00:16<00:00, 1.62it/s]\n", + "100% 26/26 [00:15<00:00, 1.65it/s]\n", + "2023-09-27 07:42:25 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:42:25 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:42:25 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:42:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:42:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:43:13 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:43:13 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:43:13 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:43:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:43:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:43:58 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:43:58 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:43:58 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:44:27 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:44:27 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:44:51 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:44:51 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:44:51 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:45:20 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:45:21 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 46/46 [00:14<00:00, 3.14it/s]\n", + "100% 46/46 [00:13<00:00, 3.34it/s]\n", + "100% 46/46 [00:12<00:00, 3.75it/s]\n", + "100% 46/46 [00:09<00:00, 4.84it/s]\n", + "100% 22/22 [00:14<00:00, 1.52it/s]\n", + "100% 22/22 [00:12<00:00, 1.70it/s]\n", + "2023-09-27 07:49:10 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:49:10 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:49:10 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:49:34 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:49:34 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:50:19 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:50:19 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:50:19 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:50:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:50:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 22/22 [00:06<00:00, 3.65it/s]\n", + "100% 22/22 [00:06<00:00, 3.65it/s]\n", + "100% 22/22 [00:04<00:00, 4.91it/s]\n", + "100% 22/22 [00:04<00:00, 4.88it/s]\n", + "100% 12/12 [00:06<00:00, 1.76it/s]\n", + "100% 12/12 [00:08<00:00, 1.45it/s]\n", + "2023-09-27 07:53:56 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", + "2023-09-27 07:53:56 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-27 07:53:56 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "gin_channels: 256 self.spk_embed_dim: 109\n", + "\n", + "2023-09-27 07:54:10 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-27 07:54:10 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n" + ] + } + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "provenance": [], + "gpuType": "T4" + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file From 5def93dd424978a00138642c95470be9998e6e79 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 27 Sep 2023 16:14:32 +0800 Subject: [PATCH 02/48] Add files via upload --- Hina_Mod_AICoverGen_colab.ipynb | 44 ++++++++++++++++----------------- 1 file changed, 22 insertions(+), 22 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index d559dd97..14e9f3af 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -6,34 +6,34 @@ "id": "39k2mOCNAh6J" }, "source": [ - "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SociallyIneptWeeb/AICoverGen/blob/main/AICoverGen_colab.ipynb)" + "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb)" ] }, { "cell_type": "markdown", + "metadata": { + "id": "YYVAKuNBc-X4" + }, "source": [ "# AICoverGen WebUI\n", "\n", "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." - ], - "metadata": { - "id": "YYVAKuNBc-X4" - } + ] }, { "cell_type": "code", "execution_count": 2, "metadata": { - "id": "vC4gLMHI9xb3", "colab": { "base_uri": "https://localhost:8080/" }, + "id": "vC4gLMHI9xb3", "outputId": "a839f0a7-e240-4020-f80a-e54fe3d27051" }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Done Cloning Repository\n" ] @@ -54,16 +54,16 @@ "cell_type": "code", "execution_count": 3, "metadata": { - "id": "odzpJHpr_PaF", "colab": { "base_uri": "https://localhost:8080/" }, + "id": "odzpJHpr_PaF", "outputId": "364aac42-6448-4752-bfff-3c32af3e49f4" }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Finsihed running this cell, proceed to the next cell\n" ] @@ -86,16 +86,16 @@ "cell_type": "code", "execution_count": 4, "metadata": { - "id": "SLWpcJc0AHSZ", "colab": { "base_uri": "https://localhost:8080/" }, + "id": "SLWpcJc0AHSZ", "outputId": "56b66089-0ae3-462a-c938-7e96e88044e5" }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Finished Downloading Voice Separation Model and Hubert Base Model\n" ] @@ -110,23 +110,18 @@ }, { "cell_type": "code", - "source": [ - "#@title Run WebUI\n", - "runpy=codecs.decode('fep/jrohv.cl','rot_13')\n", - "!python $runpy --share" - ], + "execution_count": null, "metadata": { - "id": "NEglTq6Ya9d0", "colab": { "base_uri": "https://localhost:8080/" }, + "id": "NEglTq6Ya9d0", "outputId": "617f2b65-5e54-438c-d991-a170e2ce9b7c" }, - "execution_count": null, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "2023-09-27 07:07:48.631125: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", @@ -286,14 +281,19 @@ "2023-09-27 07:54:10 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n" ] } + ], + "source": [ + "#@title Run WebUI\n", + "runpy=codecs.decode('fep/jrohv.cl','rot_13')\n", + "!python $runpy --share" ] } ], "metadata": { "accelerator": "GPU", "colab": { - "provenance": [], - "gpuType": "T4" + "gpuType": "T4", + "provenance": [] }, "kernelspec": { "display_name": "Python 3", @@ -305,4 +305,4 @@ }, "nbformat": 4, "nbformat_minor": 0 -} \ No newline at end of file +} From 87f8b7c866e49aa4f1ea8210695e0bdf10bd0f09 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 29 Sep 2023 18:43:17 +0800 Subject: [PATCH 03/48] Quick Update --- Hina_Mod_AICoverGen_colab.ipynb | 254 +++++++++++--------------------- 1 file changed, 83 insertions(+), 171 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 14e9f3af..12f84d52 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -3,10 +3,11 @@ { "cell_type": "markdown", "metadata": { - "id": "39k2mOCNAh6J" + "id": "view-in-github", + "colab_type": "text" }, "source": [ - "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb)" + "\"Open" ] }, { @@ -22,18 +23,18 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "vC4gLMHI9xb3", - "outputId": "a839f0a7-e240-4020-f80a-e54fe3d27051" + "outputId": "11624f3c-b56f-4a0b-feaa-b6667a412aed" }, "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "Done Cloning Repository\n" ] @@ -43,27 +44,41 @@ "#@title Clone repository\n", "from IPython.display import clear_output, Javascript\n", "import codecs\n", + "import threading\n", + "import time\n", "cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", - "!git clone $cloneing\n", - "%cd AICoverGen\n", + "!git clone $cloneing HRVC\n", + "def update_timer_and_print():\n", + " global timer\n", + " while True:\n", + " hours, remainder = divmod(timer, 3600)\n", + " minutes, seconds = divmod(remainder, 60)\n", + " timer_str = f'{hours:02}:{minutes:02}:{seconds:02}'\n", + " print(f'\\rTimer: {timer_str}', end='', flush=True) # Print without a newline\n", + " time.sleep(1)\n", + " timer += 1\n", + "timer = 0\n", + "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", + "\n", + "%cd HRVC\n", "clear_output()\n", "print(\"Done Cloning Repository\")" ] }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "odzpJHpr_PaF", - "outputId": "364aac42-6448-4752-bfff-3c32af3e49f4" + "outputId": "accfd3db-f230-483d-8c65-b83c0aa70c0c" }, "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "Finsihed running this cell, proceed to the next cell\n" ] @@ -84,26 +99,28 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "SLWpcJc0AHSZ", - "outputId": "56b66089-0ae3-462a-c938-7e96e88044e5" + "outputId": "6ce5cfb8-8d39-4fa0-f35f-b8cf3d58bd7f" }, "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ - "Finished Downloading Voice Separation Model and Hubert Base Model\n" + "Finished Downloading Voice Separation Model and Hubert Base Model\n", + "Timer: 00:03:38" ] } ], "source": [ "#@title Download MDXNet Vocal Separation and Hubert Base Models\n", - "!python src/download_models.py\n", + "models=codecs.decode('fep/qbjaybnq_zbqryf.cl','rot_13')\n", + "!python $models\n", "clear_output()\n", "print(\"Finished Downloading Voice Separation Model and Hubert Base Model\")" ] @@ -116,184 +133,79 @@ "base_uri": "https://localhost:8080/" }, "id": "NEglTq6Ya9d0", - "outputId": "617f2b65-5e54-438c-d991-a170e2ce9b7c" + "outputId": "46050832-09d1-4715-8b63-d1bbb8c0ae66" }, "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ - "2023-09-27 07:07:48.631125: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", + "Timer: 00:04:052023-09-29 10:28:24.893241: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", - "2023-09-27 07:07:49.538446: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", - "2023-09-27 07:07:50 | INFO | fairseq.tasks.text_to_speech | Please install tensorboardX: pip install tensorboardX\n", - "2023-09-27 07:07:51 | INFO | faiss.loader | Loading faiss with AVX2 support.\n", - "2023-09-27 07:07:51 | INFO | faiss.loader | Successfully loaded faiss with AVX2 support.\n", - "Running on local URL: http://127.0.0.1:7860\n", - "Running on public URL: https://513f29c74e8615004f.gradio.live\n", + "Timer: 00:04:062023-09-29 10:28:26.320344: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", + "Timer: 00:04:092023-09-29 10:28:28 | INFO | fairseq.tasks.text_to_speech | Please install tensorboardX: pip install tensorboardX\n", + "2023-09-29 10:28:29 | INFO | faiss.loader | Loading faiss with AVX2 support.\n", + "2023-09-29 10:28:29 | INFO | faiss.loader | Successfully loaded faiss with AVX2 support.\n", + "Timer: 00:04:10Running on local URL: http://127.0.0.1:7860\n", + "Timer: 00:04:17Running on public URL: https://e2168815069bd8c3b8.gradio.live\n", "\n", "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n", - "2023-09-27 07:08:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:08:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:08:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:08:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:09:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:09:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:09:28 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:09:28 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 38/38 [00:11<00:00, 3.17it/s]\n", - "100% 38/38 [00:10<00:00, 3.71it/s]\n", - "100% 38/38 [00:07<00:00, 5.09it/s]\n", - "100% 38/38 [00:08<00:00, 4.26it/s]\n", - "100% 20/20 [00:11<00:00, 1.74it/s]\n", - "100% 20/20 [00:13<00:00, 1.49it/s]\n", - "2023-09-27 07:13:21 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:13:21 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:13:21 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:13:55 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:13:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 38/38 [00:10<00:00, 3.66it/s]\n", - "100% 38/38 [00:12<00:00, 3.08it/s]\n", - "100% 38/38 [00:08<00:00, 4.62it/s]\n", - "100% 38/38 [00:08<00:00, 4.61it/s]\n", - "100% 20/20 [00:12<00:00, 1.59it/s]\n", - "100% 20/20 [00:11<00:00, 1.73it/s]\n", - "2023-09-27 07:18:32 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:18:32 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:18:32 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:19:05 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:19:05 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 48/48 [00:13<00:00, 3.58it/s]\n", - "100% 48/48 [00:14<00:00, 3.40it/s]\n", - "100% 48/48 [00:10<00:00, 4.80it/s]\n", - "100% 48/48 [00:11<00:00, 4.13it/s]\n", - "100% 24/24 [00:13<00:00, 1.72it/s]\n", - "100% 24/24 [00:13<00:00, 1.74it/s]\n", - "2023-09-27 07:22:40 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:22:40 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:22:40 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", + "Timer: 00:05:352023-09-29 10:29:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-29 10:29:55 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "Timer: 00:05:512023-09-29 10:30:11 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-29 10:30:11 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 40/40 [00:11<00:00, 3.34it/s]\n", + "100% 40/40 [00:13<00:00, 2.89it/s]\n", + "100% 40/40 [00:08<00:00, 4.59it/s]\n", + "100% 40/40 [00:09<00:00, 4.38it/s]\n", + "100% 20/20 [00:16<00:00, 1.24it/s]\n", + "100% 20/20 [00:11<00:00, 1.77it/s]\n", + "Timer: 00:10:322023-09-29 10:34:52 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/HRVC\n", + "2023-09-29 10:34:52 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-29 10:34:52 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "Timer: 00:10:34gin_channels: 256 self.spk_embed_dim: 109\n", "\n", - "2023-09-27 07:23:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:23:22 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 48/48 [00:13<00:00, 3.63it/s]\n", - "100% 48/48 [00:13<00:00, 3.51it/s]\n", - "100% 48/48 [00:10<00:00, 4.65it/s]\n", - "100% 48/48 [00:11<00:00, 4.12it/s]\n", + "Timer: 00:10:562023-09-29 10:35:17 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "2023-09-29 10:35:17 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "100% 50/50 [00:14<00:00, 3.55it/s]\n", + "100% 50/50 [00:15<00:00, 3.13it/s]\n", + "100% 48/48 [00:12<00:00, 3.91it/s]\n", + "100% 48/48 [00:10<00:00, 4.67it/s]\n", + "100% 24/24 [00:14<00:00, 1.61it/s]\n", "100% 24/24 [00:13<00:00, 1.76it/s]\n", - "100% 24/24 [00:13<00:00, 1.79it/s]\n", - "2023-09-27 07:32:19 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:32:19 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:32:19 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", + "Timer: 00:15:022023-09-29 10:39:23 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/HRVC\n", + "2023-09-29 10:39:23 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", + "2023-09-29 10:39:23 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", + "Timer: 00:15:04gin_channels: 256 self.spk_embed_dim: 109\n", "\n", - "2023-09-27 07:32:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:32:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 46/46 [00:12<00:00, 3.59it/s]\n", - "100% 46/46 [00:13<00:00, 3.31it/s]\n", - "100% 46/46 [00:09<00:00, 4.76it/s]\n", - "100% 46/46 [00:11<00:00, 4.01it/s]\n", - "100% 24/24 [00:13<00:00, 1.71it/s]\n", - "100% 24/24 [00:14<00:00, 1.68it/s]\n", - "2023-09-27 07:35:42 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:35:42 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:35:42 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:36:07 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:36:07 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:36:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:36:57 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:38:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:38:45 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:39:13 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:39:14 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 52/52 [00:14<00:00, 3.57it/s]\n", - "100% 52/52 [00:14<00:00, 3.53it/s]\n", - "100% 52/52 [00:12<00:00, 4.28it/s]\n", - "100% 52/52 [00:11<00:00, 4.68it/s]\n", - "100% 26/26 [00:16<00:00, 1.62it/s]\n", - "100% 26/26 [00:15<00:00, 1.65it/s]\n", - "2023-09-27 07:42:25 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:42:25 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:42:25 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:42:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:42:56 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:43:13 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:43:13 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:43:13 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:43:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:43:43 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:43:58 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:43:58 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:43:58 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:44:27 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:44:27 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:44:51 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:44:51 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:44:51 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:45:20 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:45:21 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 46/46 [00:14<00:00, 3.14it/s]\n", - "100% 46/46 [00:13<00:00, 3.34it/s]\n", - "100% 46/46 [00:12<00:00, 3.75it/s]\n", - "100% 46/46 [00:09<00:00, 4.84it/s]\n", - "100% 22/22 [00:14<00:00, 1.52it/s]\n", - "100% 22/22 [00:12<00:00, 1.70it/s]\n", - "2023-09-27 07:49:10 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:49:10 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:49:10 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:49:34 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:49:34 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:50:19 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:50:19 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:50:19 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:50:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:50:44 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 22/22 [00:06<00:00, 3.65it/s]\n", - "100% 22/22 [00:06<00:00, 3.65it/s]\n", - "100% 22/22 [00:04<00:00, 4.91it/s]\n", - "100% 22/22 [00:04<00:00, 4.88it/s]\n", - "100% 12/12 [00:06<00:00, 1.76it/s]\n", - "100% 12/12 [00:08<00:00, 1.45it/s]\n", - "2023-09-27 07:53:56 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/AICoverGen\n", - "2023-09-27 07:53:56 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-27 07:53:56 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "2023-09-27 07:54:10 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-27 07:54:10 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n" + "Timer: 00:15:322023-09-29 10:39:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", + "Timer: 00:15:332023-09-29 10:39:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", + "Timer: 00:18:54" ] } ], "source": [ "#@title Run WebUI\n", - "runpy=codecs.decode('fep/jrohv.cl','rot_13')\n", - "!python $runpy --share" + "runpice=codecs.decode('fep/jrohv.cl','rot_13')\n", + "!python $runpice --share" ] + }, + { + "cell_type": "markdown", + "source": [ + "![](https://i.pinimg.com/474x/de/72/9e/de729ecfa41b69901c42c82fff752414.jpg)" + ], + "metadata": { + "id": "npldkPdYfkrA" + } } ], "metadata": { "accelerator": "GPU", "colab": { "gpuType": "T4", - "provenance": [] + "provenance": [], + "include_colab_link": true }, "kernelspec": { "display_name": "Python 3", @@ -305,4 +217,4 @@ }, "nbformat": 4, "nbformat_minor": 0 -} +} \ No newline at end of file From 94217db81d70ef2b1576c576c6c165c52f5b9a94 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 29 Sep 2023 18:52:25 +0800 Subject: [PATCH 04/48] Created using Colaboratory --- Hina_Mod_AICoverGen_colab.ipynb | 113 ++++---------------------------- 1 file changed, 13 insertions(+), 100 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 12f84d52..b51684e0 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -23,23 +23,11 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "vC4gLMHI9xb3", - "outputId": "11624f3c-b56f-4a0b-feaa-b6667a412aed" + "id": "vC4gLMHI9xb3" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Done Cloning Repository\n" - ] - } - ], + "outputs": [], "source": [ "#@title Clone repository\n", "from IPython.display import clear_output, Javascript\n", @@ -67,23 +55,11 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "odzpJHpr_PaF", - "outputId": "accfd3db-f230-483d-8c65-b83c0aa70c0c" + "id": "odzpJHpr_PaF" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Finsihed running this cell, proceed to the next cell\n" - ] - } - ], + "outputs": [], "source": [ "#@title Install requirements\n", "!pip install -q -r requirements.txt\n", @@ -99,24 +75,11 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "SLWpcJc0AHSZ", - "outputId": "6ce5cfb8-8d39-4fa0-f35f-b8cf3d58bd7f" + "id": "SLWpcJc0AHSZ" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Finished Downloading Voice Separation Model and Hubert Base Model\n", - "Timer: 00:03:38" - ] - } - ], + "outputs": [], "source": [ "#@title Download MDXNet Vocal Separation and Hubert Base Models\n", "models=codecs.decode('fep/qbjaybnq_zbqryf.cl','rot_13')\n", @@ -129,61 +92,9 @@ "cell_type": "code", "execution_count": null, "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "NEglTq6Ya9d0", - "outputId": "46050832-09d1-4715-8b63-d1bbb8c0ae66" + "id": "NEglTq6Ya9d0" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Timer: 00:04:052023-09-29 10:28:24.893241: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", - "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", - "Timer: 00:04:062023-09-29 10:28:26.320344: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", - "Timer: 00:04:092023-09-29 10:28:28 | INFO | fairseq.tasks.text_to_speech | Please install tensorboardX: pip install tensorboardX\n", - "2023-09-29 10:28:29 | INFO | faiss.loader | Loading faiss with AVX2 support.\n", - "2023-09-29 10:28:29 | INFO | faiss.loader | Successfully loaded faiss with AVX2 support.\n", - "Timer: 00:04:10Running on local URL: http://127.0.0.1:7860\n", - "Timer: 00:04:17Running on public URL: https://e2168815069bd8c3b8.gradio.live\n", - "\n", - "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n", - "Timer: 00:05:352023-09-29 10:29:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-29 10:29:55 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "Timer: 00:05:512023-09-29 10:30:11 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-29 10:30:11 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 40/40 [00:11<00:00, 3.34it/s]\n", - "100% 40/40 [00:13<00:00, 2.89it/s]\n", - "100% 40/40 [00:08<00:00, 4.59it/s]\n", - "100% 40/40 [00:09<00:00, 4.38it/s]\n", - "100% 20/20 [00:16<00:00, 1.24it/s]\n", - "100% 20/20 [00:11<00:00, 1.77it/s]\n", - "Timer: 00:10:322023-09-29 10:34:52 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/HRVC\n", - "2023-09-29 10:34:52 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-29 10:34:52 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "Timer: 00:10:34gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "Timer: 00:10:562023-09-29 10:35:17 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "2023-09-29 10:35:17 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "100% 50/50 [00:14<00:00, 3.55it/s]\n", - "100% 50/50 [00:15<00:00, 3.13it/s]\n", - "100% 48/48 [00:12<00:00, 3.91it/s]\n", - "100% 48/48 [00:10<00:00, 4.67it/s]\n", - "100% 24/24 [00:14<00:00, 1.61it/s]\n", - "100% 24/24 [00:13<00:00, 1.76it/s]\n", - "Timer: 00:15:022023-09-29 10:39:23 | INFO | fairseq.tasks.hubert_pretraining | current directory is /content/HRVC\n", - "2023-09-29 10:39:23 | INFO | fairseq.tasks.hubert_pretraining | HubertPretrainingTask Config {'_name': 'hubert_pretraining', 'data': 'metadata', 'fine_tuning': False, 'labels': ['km'], 'label_dir': 'label', 'label_rate': 50.0, 'sample_rate': 16000, 'normalize': False, 'enable_padding': False, 'max_keep_size': None, 'max_sample_size': 250000, 'min_sample_size': 32000, 'single_target': False, 'random_crop': True, 'pad_audio': False}\n", - "2023-09-29 10:39:23 | INFO | fairseq.models.hubert.hubert | HubertModel Config: {'_name': 'hubert', 'label_rate': 50.0, 'extractor_mode': default, 'encoder_layers': 12, 'encoder_embed_dim': 768, 'encoder_ffn_embed_dim': 3072, 'encoder_attention_heads': 12, 'activation_fn': gelu, 'layer_type': transformer, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'encoder_layerdrop': 0.05, 'dropout_input': 0.1, 'dropout_features': 0.1, 'final_dim': 256, 'untie_final_proj': True, 'layer_norm_first': False, 'conv_feature_layers': '[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', 'conv_bias': False, 'logit_temp': 0.1, 'target_glu': False, 'feature_grad_mult': 0.1, 'mask_length': 10, 'mask_prob': 0.8, 'mask_selection': static, 'mask_other': 0.0, 'no_mask_overlap': False, 'mask_min_space': 1, 'mask_channel_length': 10, 'mask_channel_prob': 0.0, 'mask_channel_selection': static, 'mask_channel_other': 0.0, 'no_mask_channel_overlap': False, 'mask_channel_min_space': 1, 'conv_pos': 128, 'conv_pos_groups': 16, 'latent_temp': [2.0, 0.5, 0.999995], 'skip_masked': False, 'skip_nomask': False, 'checkpoint_activations': False, 'required_seq_len_multiple': 2, 'depthwise_conv_kernel_size': 31, 'attn_type': '', 'pos_enc_type': 'abs', 'fp16': False}\n", - "Timer: 00:15:04gin_channels: 256 self.spk_embed_dim: 109\n", - "\n", - "Timer: 00:15:322023-09-29 10:39:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/api/predict \"HTTP/1.1 200 OK\"\n", - "Timer: 00:15:332023-09-29 10:39:54 | INFO | httpx | HTTP Request: POST http://127.0.0.1:7860/reset \"HTTP/1.1 200 OK\"\n", - "Timer: 00:18:54" - ] - } - ], + "outputs": [], "source": [ "#@title Run WebUI\n", "runpice=codecs.decode('fep/jrohv.cl','rot_13')\n", @@ -205,6 +116,8 @@ "colab": { "gpuType": "T4", "provenance": [], + "private_outputs": true, + "cell_execution_strategy": "setup", "include_colab_link": true }, "kernelspec": { From 6902c4ba09248640fa701a5a851251cfccfb08ee Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 29 Sep 2023 21:40:27 +0800 Subject: [PATCH 05/48] I Forgot --- Hina_Mod_AICoverGen_colab.ipynb | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index b51684e0..5f1cae44 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -25,7 +25,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "vC4gLMHI9xb3" + "id": "vC4gLMHI9xb3", + "cellView": "form" }, "outputs": [], "source": [ @@ -57,7 +58,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "odzpJHpr_PaF" + "id": "odzpJHpr_PaF", + "cellView": "form" }, "outputs": [], "source": [ @@ -77,7 +79,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "SLWpcJc0AHSZ" + "id": "SLWpcJc0AHSZ", + "cellView": "form" }, "outputs": [], "source": [ @@ -92,7 +95,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "NEglTq6Ya9d0" + "id": "NEglTq6Ya9d0", + "cellView": "form" }, "outputs": [], "source": [ From b6b019ccb6fc6506250c0e952c42f01162535c71 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 7 Oct 2023 19:38:29 +0800 Subject: [PATCH 06/48] Reverted to use 12 pitch instead of 1 --- src/main.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/main.py b/src/main.py index dbc96769..ef3f823a 100644 --- a/src/main.py +++ b/src/main.py @@ -282,7 +282,7 @@ def song_cover_pipeline(song_input, voice_model, pitch_change, keep_files, else: orig_song_path, instrumentals_path, main_vocals_dereverb_path, backup_vocals_path = paths - pitch_change = pitch_change * 12 + pitch_change_all + pitch_change = pitch_change + pitch_change_all ai_vocals_path = os.path.join(song_dir, f'{os.path.splitext(os.path.basename(orig_song_path))[0]}_{voice_model}_p{pitch_change}_i{index_rate}_fr{filter_radius}_rms{rms_mix_rate}_pro{protect}_{f0_method}{"" if f0_method != "mangio-crepe" else f"_{crepe_hop_length}"}.wav') ai_cover_path = os.path.join(song_dir, f'{os.path.splitext(os.path.basename(orig_song_path))[0]} ({voice_model} Ver).{output_format}') @@ -320,7 +320,7 @@ def song_cover_pipeline(song_input, voice_model, pitch_change, keep_files, parser = argparse.ArgumentParser(description='Generate a AI cover song in the song_output/id directory.', add_help=True) parser.add_argument('-i', '--song-input', type=str, required=True, help='Link to a YouTube video or the filepath to a local mp3/wav file to create an AI cover of') parser.add_argument('-dir', '--rvc-dirname', type=str, required=True, help='Name of the folder in the rvc_models directory containing the RVC model file and optional index file to use') - parser.add_argument('-p', '--pitch-change', type=int, required=True, help='Change the pitch of AI Vocals only. Generally, use 1 for male to female and -1 for vice-versa. (Octaves)') + parser.add_argument('-p', '--pitch-change', type=int, required=True, help='Change the pitch of AI Vocals only. Generally, use 12 for male to female and -12 for vice-versa. (Octaves)') parser.add_argument('-k', '--keep-files', action=argparse.BooleanOptionalAction, help='Whether to keep all intermediate audio files generated in the song_output/id directory, e.g. Isolated Vocals/Instrumentals') parser.add_argument('-ir', '--index-rate', type=float, default=0.5, help='A decimal number e.g. 0.5, used to reduce/resolve the timbre leakage problem. If set to 1, more biased towards the timbre quality of the training dataset') parser.add_argument('-fr', '--filter-radius', type=int, default=3, help='A number between 0 and 7. If >=3: apply median filtering to the harvested pitch results. The value represents the filter radius and can reduce breathiness.') From 1eb57fe36dc5cea30ba7d6aca123e739d153c5af Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 7 Oct 2023 19:40:20 +0800 Subject: [PATCH 07/48] Reverted to use 12 instead of 1 on pitch --- src/webui.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/webui.py b/src/webui.py index 106997fa..2fe2be5d 100644 --- a/src/webui.py +++ b/src/webui.py @@ -190,7 +190,7 @@ def show_hop_slider(pitch_detection_algo): song_input_file.upload(process_file_upload, inputs=[song_input_file], outputs=[local_file, song_input]) with gr.Column(): - pitch = gr.Slider(-3, 3, value=0, step=1, label='Pitch Change (Vocals ONLY)', info='Generally, use 1 for male to female conversions and -1 for vice-versa. (Octaves)') + pitch = gr.Slider(-20, 20, value=0, step=1, label='Pitch Change (Vocals ONLY)', info='Generally, use 12 for male to female conversions and -12 for vice-versa. (Octaves)') pitch_all = gr.Slider(-12, 12, value=0, step=1, label='Overall Pitch Change', info='Changes pitch/key of vocals and instrumentals together. Altering this slightly reduces sound quality. (Semitones)') show_file_upload_button.click(swap_visibility, outputs=[file_upload_col, yt_link_col, song_input, local_file]) show_yt_link_button.click(swap_visibility, outputs=[yt_link_col, file_upload_col, song_input, local_file]) From 8202605a408d020765ae4e681c59d0f86025347c Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 7 Oct 2023 19:55:39 +0800 Subject: [PATCH 08/48] Reverted the pitch control to use 12 instead of 1 --- Hina_Mod_AICoverGen_colab.ipynb | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 5f1cae44..c53ee541 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -35,7 +35,7 @@ "import codecs\n", "import threading\n", "import time\n", - "cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", + "cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "!git clone $cloneing HRVC\n", "def update_timer_and_print():\n", " global timer\n", @@ -49,6 +49,7 @@ "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", "\n", + "!rm -rf sample_data\n", "%cd HRVC\n", "clear_output()\n", "print(\"Done Cloning Repository\")" From 2965e98c9134231361693cdefc560bd2f214be95 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 11 Oct 2023 22:20:35 +0800 Subject: [PATCH 09/48] Changed title(so people knows this is the modded version) --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index c53ee541..42ee6210 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -16,7 +16,7 @@ "id": "YYVAKuNBc-X4" }, "source": [ - "# AICoverGen WebUI\n", + "# AICoverGen WebUI (Modded by Hina)\n", "\n", "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." ] From 90fe6bd2f5ac175c91e9bf39c81be51a3561d97c Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 22 Oct 2023 21:43:57 +0800 Subject: [PATCH 10/48] Added random --- Hina_Mod_AICoverGen_colab.ipynb | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 42ee6210..8a211f6b 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -16,7 +16,9 @@ "id": "YYVAKuNBc-X4" }, "source": [ - "# AICoverGen WebUI (Modded by Hina)\n", + "\n", + "\n", + "# AICoverGen WebUI (Modded by [Hina](https://discordlookup.com/user/444684887363026974))\n", "\n", "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." ] @@ -109,7 +111,7 @@ { "cell_type": "markdown", "source": [ - "![](https://i.pinimg.com/474x/de/72/9e/de729ecfa41b69901c42c82fff752414.jpg)" + "[![](https://i.pinimg.com/474x/de/72/9e/de729ecfa41b69901c42c82fff752414.jpg)](https://discordlookup.com/user/444684887363026974)" ], "metadata": { "id": "npldkPdYfkrA" From e45c7d2235bd98f3c5d485a34456cc6ef2d8e89d Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 22 Oct 2023 22:35:36 +0800 Subject: [PATCH 11/48] Changed Something :D --- Hina_Mod_AICoverGen_colab.ipynb | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 8a211f6b..c1ff4330 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -38,7 +38,7 @@ "import threading\n", "import time\n", "cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", - "!git clone $cloneing HRVC\n", + "!git clone $cloneing Hina_RVC\n", "def update_timer_and_print():\n", " global timer\n", " while True:\n", @@ -52,7 +52,7 @@ "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", "\n", "!rm -rf sample_data\n", - "%cd HRVC\n", + "%cd Hina_RVC\n", "clear_output()\n", "print(\"Done Cloning Repository\")" ] From f542fb7d237282090415bc5771869be2b1cf9c10 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 01:01:40 +0800 Subject: [PATCH 12/48] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index d167389d..55d9b172 100644 --- a/README.md +++ b/README.md @@ -44,7 +44,7 @@ Then follow the instructions in the notebook to run the webui. For those without a powerful enough NVIDIA GPU, you may try AICoverGen out using Google Colab. -[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SociallyIneptWeeb/AICoverGen/blob/main/AICoverGen_colab.ipynb) +[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/hinabl/AICoverGen-Colab/blob/main/Hina_Mod_AICoverGen_colab.ipynb) For those who face issues with Google Colab notebook disconnecting after a few minutes, here's an alternative that doesn't use the WebUI. From 9e855e705b0e8595b120b867ddef185c4946657c Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 01:02:25 +0800 Subject: [PATCH 13/48] Switched back to -12 0 12 (might still be broken) --- src/main.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/main.py b/src/main.py index ef3f823a..2f08b860 100644 --- a/src/main.py +++ b/src/main.py @@ -282,7 +282,7 @@ def song_cover_pipeline(song_input, voice_model, pitch_change, keep_files, else: orig_song_path, instrumentals_path, main_vocals_dereverb_path, backup_vocals_path = paths - pitch_change = pitch_change + pitch_change_all + # pitch_change = pitch_change + pitch_change_all ai_vocals_path = os.path.join(song_dir, f'{os.path.splitext(os.path.basename(orig_song_path))[0]}_{voice_model}_p{pitch_change}_i{index_rate}_fr{filter_radius}_rms{rms_mix_rate}_pro{protect}_{f0_method}{"" if f0_method != "mangio-crepe" else f"_{crepe_hop_length}"}.wav') ai_cover_path = os.path.join(song_dir, f'{os.path.splitext(os.path.basename(orig_song_path))[0]} ({voice_model} Ver).{output_format}') From 81fb630aea448dc56b2c9277133b8461bf884b2a Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 01:14:05 +0800 Subject: [PATCH 14/48] Adding Toggle soon to either use 1 or 12 as pitch --- Hina_Mod_AICoverGen_colab.ipynb | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index c1ff4330..6fc849ad 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -27,8 +27,7 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "vC4gLMHI9xb3", - "cellView": "form" + "id": "vC4gLMHI9xb3" }, "outputs": [], "source": [ @@ -50,7 +49,30 @@ " timer += 1\n", "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", + "#=======================Auto Edit======================\n", + "\n", + "# import re\n", + "\n", + "# # Define the target string and replacement string\n", + "# target_string = \"string1ishere\"\n", + "# replacement_string = \"string2ishere\"\n", + "\n", + "# # Open the Jupyter Notebook file for reading and writing\n", + "# file_path = \"your_notebook.ipynb\"\n", + "\n", + "# with open(file_path, \"r\") as file:\n", + "# notebook_content = file.read()\n", + "\n", + "# # Use regular expressions to find and replace the target string\n", + "# notebook_content = re.sub(target_string, replacement_string, notebook_content)\n", + "\n", + "# # Write the modified content back to the Jupyter Notebook file\n", + "# with open(file_path, \"w\") as file:\n", + "# file.write(notebook_content)\n", + "\n", + "\n", "\n", + "#=====================Auto Edit End================\n", "!rm -rf sample_data\n", "%cd Hina_RVC\n", "clear_output()\n", From afc4217a59f14fe7d217872597750c8ef1df107b Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 10:57:08 +0800 Subject: [PATCH 15/48] Added Pitch Change Toggle --- Hina_Mod_AICoverGen_colab.ipynb | 42 ++++++++++++++------------------- 1 file changed, 18 insertions(+), 24 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 6fc849ad..b0eb4434 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -27,7 +27,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "vC4gLMHI9xb3" + "id": "vC4gLMHI9xb3", + "cellView": "form" }, "outputs": [], "source": [ @@ -36,7 +37,22 @@ "import codecs\n", "import threading\n", "import time\n", - "cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", + "# cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", + "\n", + "#=======================Auto Edit======================\n", + "\n", + "#@markdown ---\n", + "#@markdown Switch between ```-1 0 1``` or -12 0 12``` pitch change control\n", + "\n", + "#@markdown This can only be changed once, you need to restart the whole thing if you wanna change it again\n", + "Pitch_Change=\"12\" #@param ['1','12']\n", + "\n", + "if Pitch_Change==\"1\":\n", + " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg,'rot_13')\n", + "else:\n", + " cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", + "#=====================Auto Edit End================\n", + "\n", "!git clone $cloneing Hina_RVC\n", "def update_timer_and_print():\n", " global timer\n", @@ -49,30 +65,8 @@ " timer += 1\n", "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", - "#=======================Auto Edit======================\n", "\n", - "# import re\n", "\n", - "# # Define the target string and replacement string\n", - "# target_string = \"string1ishere\"\n", - "# replacement_string = \"string2ishere\"\n", - "\n", - "# # Open the Jupyter Notebook file for reading and writing\n", - "# file_path = \"your_notebook.ipynb\"\n", - "\n", - "# with open(file_path, \"r\") as file:\n", - "# notebook_content = file.read()\n", - "\n", - "# # Use regular expressions to find and replace the target string\n", - "# notebook_content = re.sub(target_string, replacement_string, notebook_content)\n", - "\n", - "# # Write the modified content back to the Jupyter Notebook file\n", - "# with open(file_path, \"w\") as file:\n", - "# file.write(notebook_content)\n", - "\n", - "\n", - "\n", - "#=====================Auto Edit End================\n", "!rm -rf sample_data\n", "%cd Hina_RVC\n", "clear_output()\n", From e0706eaf6177b4d3c7a42d9ec308dd03fdb2c08a Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 11:00:06 +0800 Subject: [PATCH 16/48] Fixed small Issue --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index b0eb4434..223227e8 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -48,7 +48,7 @@ "Pitch_Change=\"12\" #@param ['1','12']\n", "\n", "if Pitch_Change==\"1\":\n", - " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg,'rot_13')\n", + " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", "else:\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "#=====================Auto Edit End================\n", From ee077b78aea2bc953d121b6c7161bb4f3fb52858 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 25 Oct 2023 11:21:36 +0800 Subject: [PATCH 17/48] Ahh ye.. --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 223227e8..2bd2e7c9 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -42,7 +42,7 @@ "#=======================Auto Edit======================\n", "\n", "#@markdown ---\n", - "#@markdown Switch between ```-1 0 1``` or -12 0 12``` pitch change control\n", + "#@markdown Switch between ```-1 0 1``` or ```-12 0 12``` pitch change control\n", "\n", "#@markdown This can only be changed once, you need to restart the whole thing if you wanna change it again\n", "Pitch_Change=\"12\" #@param ['1','12']\n", From e0ebbfaae93e7a5d656088f38d5b508d975393c3 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 29 Oct 2023 19:17:27 +0800 Subject: [PATCH 18/48] Added Install To Drive --- Hina_Mod_AICoverGen_colab.ipynb | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 2bd2e7c9..a9774ce8 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -47,13 +47,30 @@ "#@markdown This can only be changed once, you need to restart the whole thing if you wanna change it again\n", "Pitch_Change=\"12\" #@param ['1','12']\n", "\n", + "#@markdown Enable if you want to install the Program to your Drive\n", "if Pitch_Change==\"1\":\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", "else:\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "#=====================Auto Edit End================\n", + "Install_To_Drive=True #@param {type:\"boolean\"}\n", + "\n", + "\n", + "#====================Use Drive============\n", + "if Install_To_Drive==True:\n", + " from google.colab import drive\n", + " drive.mount('/content/drive')\n", + "\n", + " !git clone $cloneing Hina_RVC\n", + " !mv Hina_RVC /content/drive/MyDrive/\n", + " !rm -rf sample_data\n", + " %cd /content/drive/MyDrive/Hina_RVC\n", + "else:\n", + " !git clone $cloneing Hina_RVC\n", + " !rm -rf sample_data\n", + " %cd Hina_RVC\n", + "\n", "\n", - "!git clone $cloneing Hina_RVC\n", "def update_timer_and_print():\n", " global timer\n", " while True:\n", @@ -67,8 +84,7 @@ "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", "\n", "\n", - "!rm -rf sample_data\n", - "%cd Hina_RVC\n", + "\n", "clear_output()\n", "print(\"Done Cloning Repository\")" ] From a17272dd26299a0ac68579346e124c2ce6a54610 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Mon, 30 Oct 2023 21:53:22 +0800 Subject: [PATCH 19/48] Removed Requirements that are not needed on colab --- Hina_Mod_AICoverGen_colab.ipynb | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index a9774ce8..5c4e7503 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -93,12 +93,20 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "odzpJHpr_PaF", - "cellView": "form" + "id": "odzpJHpr_PaF" }, "outputs": [], "source": [ "#@title Install requirements\n", + "!sed -i '/torch==/d' requirements.txt\n", + "!sed -i '/torchaudio==/d' requirements.txt\n", + "!sed -i '/numpy==/d' requirements.txt\n", + "!sed -i '/librosa==/d' requirements.txt\n", + "!sed -i '/Requests==/d' requirements.txt\n", + "!sed -i '/scipy==/d' requirements.txt\n", + "!sed -i '/soundfile==/d' requirements.txt\n", + "!sed -i '/tqdm==/d' requirements.txt\n", + "\n", "!pip install -q -r requirements.txt\n", "clear_output()\n", "print(\"Finished Installing Requirements\")\n", From b1615c7430b43473ee535fd9db0a3c16b5ec2ae5 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Mon, 30 Oct 2023 22:22:34 +0800 Subject: [PATCH 20/48] aforgot something... --- Hina_Mod_AICoverGen_colab.ipynb | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 5c4e7503..a50de136 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -93,7 +93,8 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "odzpJHpr_PaF" + "id": "odzpJHpr_PaF", + "cellView": "form" }, "outputs": [], "source": [ From a2faebfac310a8473c2aa9093b4f4f7aaeef5fce Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 9 Nov 2023 19:52:43 +0800 Subject: [PATCH 21/48] Update requirements.txt --- requirements.txt | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/requirements.txt b/requirements.txt index 338828fc..ca801522 100644 --- a/requirements.txt +++ b/requirements.txt @@ -2,21 +2,21 @@ deemix fairseq==0.12.2 faiss-cpu==1.7.3 ffmpeg-python>=0.2.0 -gradio==3.39.0 -lib==4.0.0 +gradio==2.5.0 +lib==2.0.0 librosa==0.9.1 -numpy==1.23.5 +numpy==1.15.0 onnxruntime_gpu praat-parselmouth>=0.4.2 -pedalboard==0.7.7 -pydub==0.25.1 -pyworld==0.3.4 -Requests==2.31.0 -scipy==1.11.1 -soundfile==0.12.1 +pedalboard==0.2.3 +pydub==0.5.1 +pyworld==0.3.1 +Requests==1.1.0 +scipy==1.4.1 +soundfile==0.9.1 --find-links https://download.pytorch.org/whl/torch_stable.html torch==2.0.1+cu118 -torchcrepe==0.0.20 -tqdm==4.65.0 +torchcrepe==0.0.10 +tqdm==4.1.0 yt_dlp==2023.7.6 -sox==1.4.1 +sox==1.2.1 From ef028c2049d14e6a5743542f59921d42d15e06e3 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 9 Nov 2023 19:54:55 +0800 Subject: [PATCH 22/48] /n --- src/configs/requirements.txt | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) create mode 100644 src/configs/requirements.txt diff --git a/src/configs/requirements.txt b/src/configs/requirements.txt new file mode 100644 index 00000000..338828fc --- /dev/null +++ b/src/configs/requirements.txt @@ -0,0 +1,22 @@ +deemix +fairseq==0.12.2 +faiss-cpu==1.7.3 +ffmpeg-python>=0.2.0 +gradio==3.39.0 +lib==4.0.0 +librosa==0.9.1 +numpy==1.23.5 +onnxruntime_gpu +praat-parselmouth>=0.4.2 +pedalboard==0.7.7 +pydub==0.25.1 +pyworld==0.3.4 +Requests==2.31.0 +scipy==1.11.1 +soundfile==0.12.1 +--find-links https://download.pytorch.org/whl/torch_stable.html +torch==2.0.1+cu118 +torchcrepe==0.0.20 +tqdm==4.65.0 +yt_dlp==2023.7.6 +sox==1.4.1 From 16e731f115c668a085aabed9ce51206fa4831fce Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 9 Nov 2023 20:54:51 +0800 Subject: [PATCH 23/48] Random --- Hina_Mod_AICoverGen_colab.ipynb | 36 ++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index a50de136..e512c9c6 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -53,7 +53,7 @@ "else:\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "#=====================Auto Edit End================\n", - "Install_To_Drive=True #@param {type:\"boolean\"}\n", + "Install_To_Drive=False\n", "\n", "\n", "#====================Use Drive============\n", @@ -99,23 +99,31 @@ "outputs": [], "source": [ "#@title Install requirements\n", - "!sed -i '/torch==/d' requirements.txt\n", - "!sed -i '/torchaudio==/d' requirements.txt\n", - "!sed -i '/numpy==/d' requirements.txt\n", - "!sed -i '/librosa==/d' requirements.txt\n", - "!sed -i '/Requests==/d' requirements.txt\n", - "!sed -i '/scipy==/d' requirements.txt\n", - "!sed -i '/soundfile==/d' requirements.txt\n", - "!sed -i '/tqdm==/d' requirements.txt\n", - "\n", - "!pip install -q -r requirements.txt\n", - "clear_output()\n", + "req=codecs.decode('erdhverzragf.gkg','rot_13')\n", + "pt=codecs.decode('fep/pbasvtf','rot_13')\n", + "\n", + "%cd $pt\n", + "time.sleep(5)\n", + "!sed -i '/torch==/d' $req\n", + "!sed -i '/torchaudio==/d' $req\n", + "!sed -i '/numpy==/d' $req\n", + "!sed -i '/librosa==/d' $req\n", + "!sed -i '/Requests==/d' $req\n", + "!sed -i '/scipy==/d' $req\n", + "!sed -i '/soundfile==/d' $req\n", + "!sed -i '/tqdm==/d' $req\n", + "\n", + "time.sleep(5)\n", + "!pip install -r requirements.txt\n", + "\n", + "%cd ../../\n", + "# clear_output()\n", "print(\"Finished Installing Requirements\")\n", "!sudo apt update\n", - "clear_output()\n", + "# clear_output()\n", "print(\"Finished Updating\")\n", "!sudo apt install sox\n", - "clear_output()\n", + "# clear_output()\n", "print(\"Finsihed running this cell, proceed to the next cell\")" ] }, From 40b55eb5f2a2f655c444a71aba39543934e63668 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 10 Nov 2023 19:39:13 +0800 Subject: [PATCH 24/48] Added Ngrok Option --- Hina_Mod_AICoverGen_colab.ipynb | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index e512c9c6..8c35861f 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -19,6 +19,9 @@ "\n", "\n", "# AICoverGen WebUI (Modded by [Hina](https://discordlookup.com/user/444684887363026974))\n", + "Original Repo by [SociallyIneptWeeb](https://github.com/SociallyIneptWeeb/AICoverGen)\n", + "\n", + "Please do credit if you do any modifications or translations to this colab notebook.\n", "\n", "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." ] @@ -38,7 +41,7 @@ "import threading\n", "import time\n", "# cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", - "\n", + "credit = \" - This Notebook was originally modified by Hina\"\n", "#=======================Auto Edit======================\n", "\n", "#@markdown ---\n", @@ -86,7 +89,7 @@ "\n", "\n", "clear_output()\n", - "print(\"Done Cloning Repository\")" + "print(\"Done Cloning Repository\"+credit)" ] }, { @@ -124,7 +127,7 @@ "print(\"Finished Updating\")\n", "!sudo apt install sox\n", "# clear_output()\n", - "print(\"Finsihed running this cell, proceed to the next cell\")" + "print(\"Finsihed running this cell, proceed to the next cell\"+credit)" ] }, { @@ -153,8 +156,21 @@ "outputs": [], "source": [ "#@title Run WebUI\n", + "#@markdown ---\n", "runpice=codecs.decode('fep/jrohv.cl','rot_13')\n", - "!python $runpice --share" + "Url=\"Gradio\" #@param ['Gradio','Ngrok']\n", + "#@markdown ---\n", + "#@markdown - Token is only needed if you are using Ngrok. You can make an account on [ngrok](https://dashboard.ngrok.com/signup) for free.\n", + "#@markdown - Click [this link](https://dashboard.ngrok.com/get-started/your-authtoken) to get your auth token, and place it here:\n", + "Token=\"\" #@param {type:\"string\"}\n", + "if Url==\"Gradio\":\n", + " !python $runpice --share\n", + "else:\n", + " ngrok.set_auth_token(Token)\n", + " ngrok.kill()\n", + " cover_tunnel = ngrok.connect(9303)\n", + " print(\"CoverGen URL:\", main_tunnel.public_url)\n", + " print(credit)" ] }, { From 94f0f6190e76a08f44af2a1ac0ad218947c77925 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 10 Nov 2023 19:58:04 +0800 Subject: [PATCH 25/48] Created using Colaboratory --- Hina_Mod_AICoverGen_colab.ipynb | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 8c35861f..cb153ddb 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -166,6 +166,8 @@ "if Url==\"Gradio\":\n", " !python $runpice --share\n", "else:\n", + " !pip install pyngrok\n", + " from pyngrok import ngrok\n", " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", " cover_tunnel = ngrok.connect(9303)\n", From 36d7541a22a3b6573cacabd79954dab43f6f7f57 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 10 Nov 2023 20:04:31 +0800 Subject: [PATCH 26/48] Created using Colaboratory --- Hina_Mod_AICoverGen_colab.ipynb | 1 + 1 file changed, 1 insertion(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index cb153ddb..7dfaf62e 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -171,6 +171,7 @@ " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", " cover_tunnel = ngrok.connect(9303)\n", + " !python $runpice --port 9303\n", " print(\"CoverGen URL:\", main_tunnel.public_url)\n", " print(credit)" ] From 767b239b7fba73e4bd936652e0570069e208e0f7 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 12 Nov 2023 13:31:11 +0800 Subject: [PATCH 27/48] Changed Top --- src/webui.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/webui.py b/src/webui.py index 2fe2be5d..bd7a6a15 100644 --- a/src/webui.py +++ b/src/webui.py @@ -168,7 +168,7 @@ def show_hop_slider(pitch_detection_algo): with gr.Blocks(title='AICoverGenWebUI') as app: - gr.Label('AICoverGen WebUI created with ❤️', show_label=False) + gr.Label('AICoverGen WebUI Modified For Colab/Kaggle ❤️', show_label=False) # main tab with gr.Tab("Generate"): From 3aaa17d71c05708d4942cf7624b05b27706c55d5 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 16 Dec 2023 09:39:27 +0800 Subject: [PATCH 28/48] Quick fix to colab updating cuda --- Hina_Mod_AICoverGen_colab.ipynb | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 7dfaf62e..d10aa0d7 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -21,6 +21,8 @@ "# AICoverGen WebUI (Modded by [Hina](https://discordlookup.com/user/444684887363026974))\n", "Original Repo by [SociallyIneptWeeb](https://github.com/SociallyIneptWeeb/AICoverGen)\n", "\n", + "Additional Credits to: [Eddy](https://github.com/Eddycrack864) and [GDR]() for finding a fix to colab update\n", + "\n", "Please do credit if you do any modifications or translations to this colab notebook.\n", "\n", "Simply click `Runtime` in the top navigation bar and `Run all`. Wait for the output of the final cell to show the public gradio url and click on it." @@ -50,7 +52,8 @@ "#@markdown This can only be changed once, you need to restart the whole thing if you wanna change it again\n", "Pitch_Change=\"12\" #@param ['1','12']\n", "\n", - "#@markdown Enable if you want to install the Program to your Drive\n", + "#@markdown This cell may take a while to install, up to 15 minutes if unlucky\n", + "##@markdown Enable if you want to install the Program to your Drive\n", "if Pitch_Change==\"1\":\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", "else:\n", @@ -87,7 +90,11 @@ "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", "\n", "\n", - "\n", + "!sudo apt update\n", + "!yes | sudo DEBIAN_FRONTEND=noninteractive apt-get -yq install cuda-11-8\n", + "clear_output()\n", + "!find / -name '*libcublasLt.so*' #Just to see the versions, you will have 11.8 and 12.2\n", + "time.sleep(10)\n", "clear_output()\n", "print(\"Done Cloning Repository\"+credit)" ] @@ -104,6 +111,7 @@ "#@title Install requirements\n", "req=codecs.decode('erdhverzragf.gkg','rot_13')\n", "pt=codecs.decode('fep/pbasvtf','rot_13')\n", + "#@markdown This cell can take up to 5 minutes to finish\n", "\n", "%cd $pt\n", "time.sleep(5)\n", @@ -126,24 +134,12 @@ "# clear_output()\n", "print(\"Finished Updating\")\n", "!sudo apt install sox\n", - "# clear_output()\n", - "print(\"Finsihed running this cell, proceed to the next cell\"+credit)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "id": "SLWpcJc0AHSZ", - "cellView": "form" - }, - "outputs": [], - "source": [ - "#@title Download MDXNet Vocal Separation and Hubert Base Models\n", + "clear_output()\n", "models=codecs.decode('fep/qbjaybnq_zbqryf.cl','rot_13')\n", "!python $models\n", "clear_output()\n", - "print(\"Finished Downloading Voice Separation Model and Hubert Base Model\")" + "print(\"Finished Downloading Voice Separation Model and Hubert Base Model\")\n", + "print(\"Finsihed running this cell, proceed to the next cell\"+credit)" ] }, { From 048bd5abfa8f447c4aef6a5ce500f08468cc358a Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 16 Dec 2023 18:29:45 +0800 Subject: [PATCH 29/48] Updated Credits --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index d10aa0d7..a9279985 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -21,7 +21,7 @@ "# AICoverGen WebUI (Modded by [Hina](https://discordlookup.com/user/444684887363026974))\n", "Original Repo by [SociallyIneptWeeb](https://github.com/SociallyIneptWeeb/AICoverGen)\n", "\n", - "Additional Credits to: [Eddy](https://github.com/Eddycrack864) and [GDR]() for finding a fix to colab update\n", + "Additional Credits to: [Eddy](https://github.com/Eddycrack864) and [GDR](https://github.com/GatienDoesStuff) for finding a fix to colab update\n", "\n", "Please do credit if you do any modifications or translations to this colab notebook.\n", "\n", From 3f8664b61d5fa858c30c069d2f574e98ea6ded53 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Tue, 2 Jan 2024 16:37:00 +0800 Subject: [PATCH 30/48] Fixed Ngrok not working --- Hina_Mod_AICoverGen_colab.ipynb | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index a9279985..9ce03c3b 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -167,9 +167,9 @@ " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", " cover_tunnel = ngrok.connect(9303)\n", - " !python $runpice --port 9303\n", - " print(\"CoverGen URL:\", main_tunnel.public_url)\n", - " print(credit)" + " print(\"CoverGen URL:\", cover_tunnel.public_url)\n", + " print(credit)\n", + " !python $runpice --listen-port 9303" ] }, { From b9a1a57d442734563b430a403445444603451175 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 18 Jan 2024 21:07:06 +0800 Subject: [PATCH 31/48] Fallback Toggle --- Hina_Mod_AICoverGen_colab.ipynb | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 9ce03c3b..97159ff8 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -52,7 +52,6 @@ "#@markdown This can only be changed once, you need to restart the whole thing if you wanna change it again\n", "Pitch_Change=\"12\" #@param ['1','12']\n", "\n", - "#@markdown This cell may take a while to install, up to 15 minutes if unlucky\n", "##@markdown Enable if you want to install the Program to your Drive\n", "if Pitch_Change==\"1\":\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/FbpvnyylVarcgJrro/NVPbireTra.tvg','rot_13')\n", @@ -89,12 +88,22 @@ "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", "\n", + "#@markdown How to use \"UseFallback\":
\n", + "#@markdown - Connect to GPU (wait to fully connect)\n", + "#@markdown - Press ``Tools`` on the menu bar\n", + "#@markdown - Press ``Command Pallete`` or ``ctrl+shift+p``\n", + "#@markdown - Search for ``Fallback`` and press ``Use fallback runtime version``\n", + "#@markdown - Check the checkbox bellow.\n", + "UseFallback=True #@param {type:\"boolean\"}\n", + "#@markdown When **NOT** using fallback -> This cell may take a while to install, up to 15 minutes if unlucky\n", + "#@markdown

``NOTE! This is a temporary fix that colab might remove when it updates again.``\n", + "if UseFallback==False:\n", + " !sudo apt update\n", + " !yes | sudo DEBIAN_FRONTEND=noninteractive apt-get -yq install cuda-11-8\n", + " clear_output()\n", + " !find / -name '*libcublasLt.so*' #Just to see the versions, you will have 11.8 and 12.2\n", + " time.sleep(10)\n", "\n", - "!sudo apt update\n", - "!yes | sudo DEBIAN_FRONTEND=noninteractive apt-get -yq install cuda-11-8\n", - "clear_output()\n", - "!find / -name '*libcublasLt.so*' #Just to see the versions, you will have 11.8 and 12.2\n", - "time.sleep(10)\n", "clear_output()\n", "print(\"Done Cloning Repository\"+credit)" ] From edfc3e9fd81b766c7c4fd29397e9c1989eecf60b Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 19 Jan 2024 18:07:03 +0800 Subject: [PATCH 32/48] Set UseFallback to default False --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 97159ff8..13fe7f23 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -94,7 +94,7 @@ "#@markdown - Press ``Command Pallete`` or ``ctrl+shift+p``\n", "#@markdown - Search for ``Fallback`` and press ``Use fallback runtime version``\n", "#@markdown - Check the checkbox bellow.\n", - "UseFallback=True #@param {type:\"boolean\"}\n", + "UseFallback=False #@param {type:\"boolean\"}\n", "#@markdown When **NOT** using fallback -> This cell may take a while to install, up to 15 minutes if unlucky\n", "#@markdown

``NOTE! This is a temporary fix that colab might remove when it updates again.``\n", "if UseFallback==False:\n", From db8a752ac72b928cfe6c73763283c5aff3983086 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 19 Jan 2024 20:47:55 +0800 Subject: [PATCH 33/48] Saved Backup --- src/{webui.py => webui_old.py} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename src/{webui.py => webui_old.py} (100%) diff --git a/src/webui.py b/src/webui_old.py similarity index 100% rename from src/webui.py rename to src/webui_old.py From f7d6fbd7a80f1414fdc2b70a6f9516f35b0fe4b4 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 19 Jan 2024 20:48:47 +0800 Subject: [PATCH 34/48] Temp Fix to Gradio on Colab --- src/webui.py | 322 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 322 insertions(+) create mode 100644 src/webui.py diff --git a/src/webui.py b/src/webui.py new file mode 100644 index 00000000..67b44975 --- /dev/null +++ b/src/webui.py @@ -0,0 +1,322 @@ +import json +import os +import shutil +import urllib.request +import zipfile +from argparse import ArgumentParser + +import gradio as gr + +from main import song_cover_pipeline + +BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) + +mdxnet_models_dir = os.path.join(BASE_DIR, 'mdxnet_models') +rvc_models_dir = os.path.join(BASE_DIR, 'rvc_models') +output_dir = os.path.join(BASE_DIR, 'song_output') + + +def get_current_models(models_dir): + models_list = os.listdir(models_dir) + items_to_remove = ['hubert_base.pt', 'MODELS.txt', 'public_models.json', 'rmvpe.pt'] + return [item for item in models_list if item not in items_to_remove] + + +def update_models_list(): + models_l = get_current_models(rvc_models_dir) + return gr.Dropdown.update(choices=models_l) + + +def load_public_models(): + models_table = [] + for model in public_models['voice_models']: + if not model['name'] in voice_models: + model = [model['name'], model['description'], model['credit'], model['url'], ', '.join(model['tags'])] + models_table.append(model) + + tags = list(public_models['tags'].keys()) + return gr.DataFrame.update(value=models_table), gr.CheckboxGroup.update(choices=tags) + + +def extract_zip(extraction_folder, zip_name): + os.makedirs(extraction_folder) + with zipfile.ZipFile(zip_name, 'r') as zip_ref: + zip_ref.extractall(extraction_folder) + os.remove(zip_name) + + index_filepath, model_filepath = None, None + for root, dirs, files in os.walk(extraction_folder): + for name in files: + if name.endswith('.index') and os.stat(os.path.join(root, name)).st_size > 1024 * 100: + index_filepath = os.path.join(root, name) + + if name.endswith('.pth') and os.stat(os.path.join(root, name)).st_size > 1024 * 1024 * 40: + model_filepath = os.path.join(root, name) + + if not model_filepath: + raise gr.Error(f'No .pth model file was found in the extracted zip. Please check {extraction_folder}.') + + # move model and index file to extraction folder + os.rename(model_filepath, os.path.join(extraction_folder, os.path.basename(model_filepath))) + if index_filepath: + os.rename(index_filepath, os.path.join(extraction_folder, os.path.basename(index_filepath))) + + # remove any unnecessary nested folders + for filepath in os.listdir(extraction_folder): + if os.path.isdir(os.path.join(extraction_folder, filepath)): + shutil.rmtree(os.path.join(extraction_folder, filepath)) + + +def download_online_model(url, dir_name, progress=gr.Progress()): + try: + progress(0, desc=f'[~] Downloading voice model with name {dir_name}...') + zip_name = url.split('/')[-1] + extraction_folder = os.path.join(rvc_models_dir, dir_name) + if os.path.exists(extraction_folder): + raise gr.Error(f'Voice model directory {dir_name} already exists! Choose a different name for your voice model.') + + if 'pixeldrain.com' in url: + url = f'https://pixeldrain.com/api/file/{zip_name}' + + urllib.request.urlretrieve(url, zip_name) + + progress(0.5, desc='[~] Extracting zip...') + extract_zip(extraction_folder, zip_name) + return f'[+] {dir_name} Model successfully downloaded!' + + except Exception as e: + raise gr.Error(str(e)) + + +def upload_local_model(zip_path, dir_name, progress=gr.Progress()): + try: + extraction_folder = os.path.join(rvc_models_dir, dir_name) + if os.path.exists(extraction_folder): + raise gr.Error(f'Voice model directory {dir_name} already exists! Choose a different name for your voice model.') + + zip_name = zip_path.name + progress(0.5, desc='[~] Extracting zip...') + extract_zip(extraction_folder, zip_name) + return f'[+] {dir_name} Model successfully uploaded!' + + except Exception as e: + raise gr.Error(str(e)) + + +def filter_models(tags, query): + models_table = [] + + # no filter + if len(tags) == 0 and len(query) == 0: + for model in public_models['voice_models']: + models_table.append([model['name'], model['description'], model['credit'], model['url'], model['tags']]) + + # filter based on tags and query + elif len(tags) > 0 and len(query) > 0: + for model in public_models['voice_models']: + if all(tag in model['tags'] for tag in tags): + model_attributes = f"{model['name']} {model['description']} {model['credit']} {' '.join(model['tags'])}".lower() + if query.lower() in model_attributes: + models_table.append([model['name'], model['description'], model['credit'], model['url'], model['tags']]) + + # filter based on only tags + elif len(tags) > 0: + for model in public_models['voice_models']: + if all(tag in model['tags'] for tag in tags): + models_table.append([model['name'], model['description'], model['credit'], model['url'], model['tags']]) + + # filter based on only query + else: + for model in public_models['voice_models']: + model_attributes = f"{model['name']} {model['description']} {model['credit']} {' '.join(model['tags'])}".lower() + if query.lower() in model_attributes: + models_table.append([model['name'], model['description'], model['credit'], model['url'], model['tags']]) + + return gr.DataFrame.update(value=models_table) + + +def pub_dl_autofill(pub_models, event: gr.SelectData): + return gr.Text.update(value=pub_models.loc[event.index[0], 'URL']), gr.Text.update(value=pub_models.loc[event.index[0], 'Model Name']) + + +def swap_visibility(): + return gr.update(visible=True), gr.update(visible=False), gr.update(value=''), gr.update(value=None) + + +def process_file_upload(file): + return file.name, gr.update(value=file.name) + + +def show_hop_slider(pitch_detection_algo): + if pitch_detection_algo == 'mangio-crepe': + return gr.update(visible=True) + else: + return gr.update(visible=False) + + +if __name__ == '__main__': + parser = ArgumentParser(description='Generate a AI cover song in the song_output/id directory.', add_help=True) + parser.add_argument("--share", action="store_true", dest="share_enabled", default=False, help="Enable sharing") + parser.add_argument("--listen", action="store_true", default=False, help="Make the WebUI reachable from your local network.") + parser.add_argument('--listen-host', type=str, help='The hostname that the server will use.') + parser.add_argument('--listen-port', type=int, help='The listening port that the server will use.') + args = parser.parse_args() + + voice_models = get_current_models(rvc_models_dir) + with open(os.path.join(rvc_models_dir, 'public_models.json'), encoding='utf8') as infile: + public_models = json.load(infile) + + with gr.Blocks(title='AICoverGenWebUI') as app: + + gr.Label('AICoverGen WebUI Modified For Colab/Kaggle ❤️', show_label=False) + + # main tab + with gr.Tab("Generate"): + + with gr.Accordion('Main Options'): + with gr.Row(): + with gr.Column(): + rvc_model = gr.Dropdown(voice_models, label='Voice Models', info='Models folder "AICoverGen --> rvc_models". After new models are added into this folder, click the refresh button') + ref_btn = gr.Button('Refresh Models 🔁', variant='primary') + + with gr.Column() as yt_link_col: + song_input = gr.Text(label='Song input', info='Link to a song on YouTube or full path to a local file. For file upload, click the button below.') + show_file_upload_button = gr.Button('Upload file instead') + + with gr.Column(visible=False) as file_upload_col: + local_file = gr.File(label='Audio file') + song_input_file = gr.UploadButton('Upload 📂', file_types=['audio'], variant='primary') + show_yt_link_button = gr.Button('Paste YouTube link/Path to local file instead') + song_input_file.upload(process_file_upload, inputs=[song_input_file], outputs=[local_file, song_input]) + + with gr.Column(): + pitch = gr.Slider(-20, 20, value=0, step=1, label='Pitch Change (Vocals ONLY)', info='Generally, use 12 for male to female conversions and -12 for vice-versa. (Octaves)') + pitch_all = gr.Slider(-12, 12, value=0, step=1, label='Overall Pitch Change', info='Changes pitch/key of vocals and instrumentals together. Altering this slightly reduces sound quality. (Semitones)') + show_file_upload_button.click(swap_visibility, outputs=[file_upload_col, yt_link_col, song_input, local_file]) + show_yt_link_button.click(swap_visibility, outputs=[yt_link_col, file_upload_col, song_input, local_file]) + + with gr.Accordion('Voice conversion options', open=False): + with gr.Row(): + index_rate = gr.Slider(0, 1, value=0.5, label='Index Rate', info="Controls how much of the AI voice's accent to keep in the vocals") + filter_radius = gr.Slider(0, 7, value=3, step=1, label='Filter radius', info='If >=3: apply median filtering median filtering to the harvested pitch results. Can reduce breathiness') + rms_mix_rate = gr.Slider(0, 1, value=0.25, label='RMS mix rate', info="Control how much to mimic the original vocal's loudness (0) or a fixed loudness (1)") + protect = gr.Slider(0, 0.5, value=0.33, label='Protect rate', info='Protect voiceless consonants and breath sounds. Set to 0.5 to disable.') + with gr.Column(): + f0_method = gr.Dropdown(['rmvpe', 'mangio-crepe'], value='rmvpe', label='Pitch detection algorithm', info='Best option is rmvpe (clarity in vocals), then mangio-crepe (smoother vocals)') + crepe_hop_length = gr.Slider(32, 320, value=128, step=1, visible=False, label='Crepe hop length', info='Lower values leads to longer conversions and higher risk of voice cracks, but better pitch accuracy.') + f0_method.change(show_hop_slider, inputs=f0_method, outputs=crepe_hop_length) + keep_files = gr.Checkbox(label='Keep intermediate files', info='Keep all audio files generated in the song_output/id directory, e.g. Isolated Vocals/Instrumentals. Leave unchecked to save space') + + with gr.Accordion('Audio mixing options', open=False): + gr.Markdown('### Volume Change (decibels)') + with gr.Row(): + main_gain = gr.Slider(-20, 20, value=0, step=1, label='Main Vocals') + backup_gain = gr.Slider(-20, 20, value=0, step=1, label='Backup Vocals') + inst_gain = gr.Slider(-20, 20, value=0, step=1, label='Music') + + gr.Markdown('### Reverb Control on AI Vocals') + with gr.Row(): + reverb_rm_size = gr.Slider(0, 1, value=0.15, label='Room size', info='The larger the room, the longer the reverb time') + reverb_wet = gr.Slider(0, 1, value=0.2, label='Wetness level', info='Level of AI vocals with reverb') + reverb_dry = gr.Slider(0, 1, value=0.8, label='Dryness level', info='Level of AI vocals without reverb') + reverb_damping = gr.Slider(0, 1, value=0.7, label='Damping level', info='Absorption of high frequencies in the reverb') + + gr.Markdown('### Audio Output Format') + output_format = gr.Dropdown(['mp3', 'wav'], value='mp3', label='Output file type', info='mp3: small file size, decent quality. wav: Large file size, best quality') + + with gr.Row(): + clear_btn = gr.ClearButton(value='Clear', components=[song_input, rvc_model, keep_files, local_file]) + generate_btn = gr.Button("Generate", variant='primary') + ai_cover = gr.Audio(label='AI Cover', show_share_button=False) + + ref_btn.click(update_models_list, None, outputs=rvc_model) + is_webui = gr.Number(value=1, visible=False) + generate_btn.click(song_cover_pipeline, + inputs=[song_input, rvc_model, pitch, keep_files, is_webui, main_gain, backup_gain, + inst_gain, index_rate, filter_radius, rms_mix_rate, f0_method, crepe_hop_length, + protect, pitch_all, reverb_rm_size, reverb_wet, reverb_dry, reverb_damping, + output_format], + outputs=[ai_cover]) + clear_btn.click(lambda: [0, 0, 0, 0, 0.5, 3, 0.25, 0.33, 'rmvpe', 128, 0, 0.15, 0.2, 0.8, 0.7, 'mp3', None], + outputs=[pitch, main_gain, backup_gain, inst_gain, index_rate, filter_radius, rms_mix_rate, + protect, f0_method, crepe_hop_length, pitch_all, reverb_rm_size, reverb_wet, + reverb_dry, reverb_damping, output_format, ai_cover]) + + # Download tab + with gr.Tab('Download model'): + + with gr.Tab('From HuggingFace/Pixeldrain URL'): + with gr.Row(): + model_zip_link = gr.Text(label='Download link to model', info='Should be a zip file containing a .pth model file and an optional .index file.') + model_name = gr.Text(label='Name your model', info='Give your new model a unique name from your other voice models.') + + with gr.Row(): + download_btn = gr.Button('Download 🌐', variant='primary', scale=19) + dl_output_message = gr.Text(label='Output Message', interactive=False, scale=20) + + download_btn.click(download_online_model, inputs=[model_zip_link, model_name], outputs=dl_output_message) + + gr.Markdown('## Input Examples') + gr.Examples( + [ + ['https://huggingface.co/phant0m4r/LiSA/resolve/main/LiSA.zip', 'Lisa'], + ['https://pixeldrain.com/u/3tJmABXA', 'Gura'], + ['https://huggingface.co/Kit-Lemonfoot/kitlemonfoot_rvc_models/resolve/main/AZKi%20(Hybrid).zip', 'Azki'] + ], + [model_zip_link, model_name], + [], + download_online_model, + ) + + with gr.Tab('From Public Index'): + + gr.Markdown('## How to use') + gr.Markdown('- Click Initialize public models table') + gr.Markdown('- Filter models using tags or search bar') + gr.Markdown('- Select a row to autofill the download link and model name') + gr.Markdown('- Click Download') + + with gr.Row(): + pub_zip_link = gr.Text(label='Download link to model') + pub_model_name = gr.Text(label='Model name') + + with gr.Row(): + download_pub_btn = gr.Button('Download 🌐', variant='primary', scale=19) + pub_dl_output_message = gr.Text(label='Output Message', interactive=False, scale=20) + + filter_tags = gr.CheckboxGroup(value=[], label='Show voice models with tags', choices=[]) + search_query = gr.Text(label='Search') + load_public_models_button = gr.Button(value='Initialize public models table', variant='primary') + + public_models_table = gr.DataFrame(value=[], headers=['Model Name', 'Description', 'Credit', 'URL', 'Tags'], label='Available Public Models', interactive=False) + public_models_table.select(pub_dl_autofill, inputs=[public_models_table], outputs=[pub_zip_link, pub_model_name]) + load_public_models_button.click(load_public_models, outputs=[public_models_table, filter_tags]) + search_query.change(filter_models, inputs=[filter_tags, search_query], outputs=public_models_table) + filter_tags.change(filter_models, inputs=[filter_tags, search_query], outputs=public_models_table) + download_pub_btn.click(download_online_model, inputs=[pub_zip_link, pub_model_name], outputs=pub_dl_output_message) + + # Upload tab + with gr.Tab('Upload model'): + gr.Markdown('## Upload locally trained RVC v2 model and index file') + gr.Markdown('- Find model file (weights folder) and optional index file (logs/[name] folder)') + gr.Markdown('- Compress files into zip file') + gr.Markdown('- Upload zip file and give unique name for voice') + gr.Markdown('- Click Upload model') + + with gr.Row(): + with gr.Column(): + zip_file = gr.File(label='Zip file') + + local_model_name = gr.Text(label='Model name') + + with gr.Row(): + model_upload_button = gr.Button('Upload model', variant='primary', scale=19) + local_upload_output_message = gr.Text(label='Output Message', interactive=False, scale=20) + model_upload_button.click(upload_local_model, inputs=[zip_file, local_model_name], outputs=local_upload_output_message) + + app.launch( + share=args.share_enabled, + enable_queue=True, + server_name=None if not args.listen else (args.listen_host or '0.0.0.0'), + server_port=9999, + ) From 0b4f6de8f40a8c9fb81743a40cb4ca53a49bebd9 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 21 Jan 2024 21:06:39 +0800 Subject: [PATCH 35/48] Fixed Ngrok Port --- Hina_Mod_AICoverGen_colab.ipynb | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 13fe7f23..0b7793d9 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -94,7 +94,7 @@ "#@markdown - Press ``Command Pallete`` or ``ctrl+shift+p``\n", "#@markdown - Search for ``Fallback`` and press ``Use fallback runtime version``\n", "#@markdown - Check the checkbox bellow.\n", - "UseFallback=False #@param {type:\"boolean\"}\n", + "UseFallback=True #@param {type:\"boolean\"}\n", "#@markdown When **NOT** using fallback -> This cell may take a while to install, up to 15 minutes if unlucky\n", "#@markdown

``NOTE! This is a temporary fix that colab might remove when it updates again.``\n", "if UseFallback==False:\n", @@ -175,10 +175,10 @@ " from pyngrok import ngrok\n", " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", - " cover_tunnel = ngrok.connect(9303)\n", + " cover_tunnel = ngrok.connect(9999)\n", " print(\"CoverGen URL:\", cover_tunnel.public_url)\n", " print(credit)\n", - " !python $runpice --listen-port 9303" + " !python $runpice --listen-port 9999" ] }, { From 153b8d59af625c2e53425e09882a4f1e0faf5943 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Mon, 22 Jan 2024 23:33:56 +0800 Subject: [PATCH 36/48] Set UseFallback to False on default --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 0b7793d9..7a0bfc58 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -94,7 +94,7 @@ "#@markdown - Press ``Command Pallete`` or ``ctrl+shift+p``\n", "#@markdown - Search for ``Fallback`` and press ``Use fallback runtime version``\n", "#@markdown - Check the checkbox bellow.\n", - "UseFallback=True #@param {type:\"boolean\"}\n", + "UseFallback=False #@param {type:\"boolean\"}\n", "#@markdown When **NOT** using fallback -> This cell may take a while to install, up to 15 minutes if unlucky\n", "#@markdown

``NOTE! This is a temporary fix that colab might remove when it updates again.``\n", "if UseFallback==False:\n", From b93f81ecd366939ddc0e6f6ac0c3aaac96d0d542 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sat, 3 Feb 2024 23:41:36 +0800 Subject: [PATCH 37/48] FINALLY! Fixed cuda error --- Hina_Mod_AICoverGen_colab.ipynb | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 7a0bfc58..51b1728b 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -87,17 +87,11 @@ " timer += 1\n", "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", - "\n", - "#@markdown How to use \"UseFallback\":
\n", - "#@markdown - Connect to GPU (wait to fully connect)\n", - "#@markdown - Press ``Tools`` on the menu bar\n", - "#@markdown - Press ``Command Pallete`` or ``ctrl+shift+p``\n", - "#@markdown - Search for ``Fallback`` and press ``Use fallback runtime version``\n", - "#@markdown - Check the checkbox bellow.\n", - "UseFallback=False #@param {type:\"boolean\"}\n", - "#@markdown When **NOT** using fallback -> This cell may take a while to install, up to 15 minutes if unlucky\n", - "#@markdown

``NOTE! This is a temporary fix that colab might remove when it updates again.``\n", - "if UseFallback==False:\n", + "#@markdown ---\n", + "LongInstall=False #@param {type:\"boolean\"}\n", + "#@markdown When using LongInstall -> This cell may take a while to install, up to 15 minutes if unlucky\n", + "#@markdown

``NOTE! The L̲o̲n̲g̲I̲n̲s̲t̲a̲l̲l̲ toggle is only needed when it fails to use the GPU when making a cover``\n", + "if LongInstall==True:\n", " !sudo apt update\n", " !yes | sudo DEBIAN_FRONTEND=noninteractive apt-get -yq install cuda-11-8\n", " clear_output()\n", @@ -135,7 +129,8 @@ "\n", "time.sleep(5)\n", "!pip install -r requirements.txt\n", - "\n", + "# install cuda fix\n", + "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", "%cd ../../\n", "# clear_output()\n", "print(\"Finished Installing Requirements\")\n", From 3e574c4cec952c7c8c59f05baa82462289788ade Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Tue, 6 Feb 2024 21:43:52 +0800 Subject: [PATCH 38/48] Added Cloudflared option on URL --- Hina_Mod_AICoverGen_colab.ipynb | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 51b1728b..3b0b039a 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -82,7 +82,7 @@ " hours, remainder = divmod(timer, 3600)\n", " minutes, seconds = divmod(remainder, 60)\n", " timer_str = f'{hours:02}:{minutes:02}:{seconds:02}'\n", - " print(f'\\rTimer: {timer_str}', end='', flush=True) # Print without a newline\n", + " print(f'\\rTimer: {timer_str} ', end='', flush=True) # Print without a newline\n", " time.sleep(1)\n", " timer += 1\n", "timer = 0\n", @@ -158,13 +158,27 @@ "#@title Run WebUI\n", "#@markdown ---\n", "runpice=codecs.decode('fep/jrohv.cl','rot_13')\n", - "Url=\"Gradio\" #@param ['Gradio','Ngrok']\n", + "Url=\"Cloudflared\" #@param ['Gradio','Ngrok','Cloudflared']\n", "#@markdown ---\n", "#@markdown - Token is only needed if you are using Ngrok. You can make an account on [ngrok](https://dashboard.ngrok.com/signup) for free.\n", "#@markdown - Click [this link](https://dashboard.ngrok.com/get-started/your-authtoken) to get your auth token, and place it here:\n", "Token=\"\" #@param {type:\"string\"}\n", "if Url==\"Gradio\":\n", " !python $runpice --share\n", + "elif Url == \"Cloudflared\":\n", + " # Install Cloudflared\n", + " !curl -LO https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64.deb\n", + " !dpkg -i cloudflared-linux-amd64.deb\n", + " # Delete log file\n", + " !rm -rf nohup.out\n", + " # Run cloudflared\n", + " !nohup cloudflared tunnel --url localhost:9999 &\n", + " clear_output()\n", + "\n", + " time.sleep(5)\n", + "\n", + " !grep -oE \"https://[a-zA-Z0-9.-]+\\.trycloudflare\\.com\" nohup.out\n", + " !python $runpice --listen-port 9999\n", "else:\n", " !pip install pyngrok\n", " from pyngrok import ngrok\n", From e600712b5001bd8fdbcff82c87cf9481fb6051f0 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Wed, 7 Feb 2024 19:23:28 +0800 Subject: [PATCH 39/48] Updated Gradio Client --- Hina_Mod_AICoverGen_colab.ipynb | 1 + 1 file changed, 1 insertion(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 3b0b039a..b7c47d56 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -129,6 +129,7 @@ "\n", "time.sleep(5)\n", "!pip install -r requirements.txt\n", + "!pip install gradio-client==0.8.1\n", "# install cuda fix\n", "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", "%cd ../../\n", From 87789c25b12f5e3b2a028c886e8a53608b7e3d7b Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 9 Feb 2024 18:28:29 +0800 Subject: [PATCH 40/48] Updated gradio --- Hina_Mod_AICoverGen_colab.ipynb | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index b7c47d56..8d86a83e 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -129,7 +129,8 @@ "\n", "time.sleep(5)\n", "!pip install -r requirements.txt\n", - "!pip install gradio-client==0.8.1\n", + "# !pip install gradio-client==0.8.1\n", + "!pip install gradio==3.48.0\n", "# install cuda fix\n", "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", "%cd ../../\n", From c4237467195d98749ae707ac7b4ace2404f2c488 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 15 Feb 2024 20:00:51 +0800 Subject: [PATCH 41/48] changed np.int to int --- Hina_Mod_AICoverGen_colab.ipynb | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 8d86a83e..d21d0bcd 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -58,6 +58,7 @@ "else:\n", " cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "#=====================Auto Edit End================\n", + "# this doesnt work btw :D\n", "Install_To_Drive=False\n", "\n", "\n", @@ -127,9 +128,11 @@ "!sed -i '/soundfile==/d' $req\n", "!sed -i '/tqdm==/d' $req\n", "\n", + "# Fix Numpy deprecation\n", + "!sed -i 's/(np.int)/(int)/g' src/vc_infer_pipeline.py\n", + "\n", "time.sleep(5)\n", "!pip install -r requirements.txt\n", - "# !pip install gradio-client==0.8.1\n", "!pip install gradio==3.48.0\n", "# install cuda fix\n", "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", @@ -160,7 +163,7 @@ "#@title Run WebUI\n", "#@markdown ---\n", "runpice=codecs.decode('fep/jrohv.cl','rot_13')\n", - "Url=\"Cloudflared\" #@param ['Gradio','Ngrok','Cloudflared']\n", + "Url=\"Gradio\" #@param ['Gradio','Ngrok','Cloudflared']\n", "#@markdown ---\n", "#@markdown - Token is only needed if you are using Ngrok. You can make an account on [ngrok](https://dashboard.ngrok.com/signup) for free.\n", "#@markdown - Click [this link](https://dashboard.ngrok.com/get-started/your-authtoken) to get your auth token, and place it here:\n", From 55caa4e1c91c8ece22392985308c3701eb19d88b Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 15 Feb 2024 23:47:34 +0800 Subject: [PATCH 42/48] RE: Fixed wrong path --- Hina_Mod_AICoverGen_colab.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index d21d0bcd..89bd6a39 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -129,7 +129,7 @@ "!sed -i '/tqdm==/d' $req\n", "\n", "# Fix Numpy deprecation\n", - "!sed -i 's/(np.int)/(int)/g' src/vc_infer_pipeline.py\n", + "!sed -i 's/(np.int)/(int)/g' '../vc_infer_pipeline.py'\n", "\n", "time.sleep(5)\n", "!pip install -r requirements.txt\n", From e6ccbc92307fa5a7f6b158bd92fcaaa315066495 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Tue, 20 Feb 2024 00:09:29 +0800 Subject: [PATCH 43/48] Updated to work with Pitch 1 --- Hina_Mod_AICoverGen_colab.ipynb | 54 ++++++++++++++++++++++----------- 1 file changed, 37 insertions(+), 17 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 89bd6a39..4d5fcc48 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -74,6 +74,7 @@ "else:\n", " !git clone $cloneing Hina_RVC\n", " !rm -rf sample_data\n", + " %cd /content/\n", " %cd Hina_RVC\n", "\n", "\n", @@ -116,27 +117,46 @@ "req=codecs.decode('erdhverzragf.gkg','rot_13')\n", "pt=codecs.decode('fep/pbasvtf','rot_13')\n", "#@markdown This cell can take up to 5 minutes to finish\n", + "if Pitch_Change==\"12\":\n", + " print(\"Using Pitch 12\")\n", + " %cd $pt\n", + " time.sleep(5)\n", + " !sed -i '/torch==/d' $req\n", + " !sed -i '/torchaudio==/d' $req\n", + " !sed -i '/numpy==/d' $req\n", + " !sed -i '/librosa==/d' $req\n", + " !sed -i '/Requests==/d' $req\n", + " !sed -i '/scipy==/d' $req\n", + " !sed -i '/soundfile==/d' $req\n", + " !sed -i '/tqdm==/d' $req\n", + "\n", + " time.sleep(5)\n", + " !pip install -r requirements.txt\n", + " # Fix Numpy deprecation\n", + " !sed -i 's/(np.int)/(int)/g' '../vc_infer_pipeline.py'\n", + " %cd ../../\n", + "else:\n", + " print(\"Using Pitch 1\")\n", + " time.sleep(5)\n", + " !sed -i '/torch==/d' $req\n", + " !sed -i '/torchaudio==/d' $req\n", + " !sed -i '/numpy==/d' $req\n", + " !sed -i '/librosa==/d' $req\n", + " !sed -i '/Requests==/d' $req\n", + " !sed -i '/scipy==/d' $req\n", + " !sed -i '/soundfile==/d' $req\n", + " !sed -i '/tqdm==/d' $req\n", "\n", - "%cd $pt\n", - "time.sleep(5)\n", - "!sed -i '/torch==/d' $req\n", - "!sed -i '/torchaudio==/d' $req\n", - "!sed -i '/numpy==/d' $req\n", - "!sed -i '/librosa==/d' $req\n", - "!sed -i '/Requests==/d' $req\n", - "!sed -i '/scipy==/d' $req\n", - "!sed -i '/soundfile==/d' $req\n", - "!sed -i '/tqdm==/d' $req\n", - "\n", - "# Fix Numpy deprecation\n", - "!sed -i 's/(np.int)/(int)/g' '../vc_infer_pipeline.py'\n", - "\n", - "time.sleep(5)\n", - "!pip install -r requirements.txt\n", + " time.sleep(5)\n", + " !pip install -r requirements.txt\n", + " # Fix Numpy deprecation\n", + " !sed -i 's/(np.int)/(int)/g' 'src/vc_infer_pipeline.py'\n", "!pip install gradio==3.48.0\n", + "\n", + "\n", + "\n", "# install cuda fix\n", "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", - "%cd ../../\n", "# clear_output()\n", "print(\"Finished Installing Requirements\")\n", "!sudo apt update\n", From 3fb2148ba33dd0b9238f651fea107d37be3e5829 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Sun, 16 Jun 2024 19:18:33 +0800 Subject: [PATCH 44/48] import os.. for reasons --- Hina_Mod_AICoverGen_colab.ipynb | 1 + 1 file changed, 1 insertion(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 4d5fcc48..03c843e4 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -42,6 +42,7 @@ "import codecs\n", "import threading\n", "import time\n", + "import os # for reasons...\n", "# cloneing=codecs.decode('uggcf://tvguho.pbz/uvanoy/NVPbireTra-Pbyno.tvg','rot_13')\n", "credit = \" - This Notebook was originally modified by Hina\"\n", "#=======================Auto Edit======================\n", From 6b80ba484ce7da28664ff4670989decdfbcbedc6 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Thu, 25 Jul 2024 15:02:49 +0800 Subject: [PATCH 45/48] Updated pip --- Hina_Mod_AICoverGen_colab.ipynb | 1 + 1 file changed, 1 insertion(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 03c843e4..9470fac1 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -118,6 +118,7 @@ "req=codecs.decode('erdhverzragf.gkg','rot_13')\n", "pt=codecs.decode('fep/pbasvtf','rot_13')\n", "#@markdown This cell can take up to 5 minutes to finish\n", + "!pip install pip==23.3.1\n", "if Pitch_Change==\"12\":\n", " print(\"Using Pitch 12\")\n", " %cd $pt\n", From 0d5f8ec00b1ab7b158010a411a3961b501a99217 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 17 Jan 2025 22:18:07 +0800 Subject: [PATCH 46/48] Downgrade --- Hina_Mod_AICoverGen_colab.ipynb | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 9470fac1..131e9a6e 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -90,6 +90,7 @@ " timer += 1\n", "timer = 0\n", "threading.Thread(target=update_timer_and_print, daemon=True).start()\n", + "\n", "#@markdown ---\n", "LongInstall=False #@param {type:\"boolean\"}\n", "#@markdown When using LongInstall -> This cell may take a while to install, up to 15 minutes if unlucky\n", @@ -118,7 +119,14 @@ "req=codecs.decode('erdhverzragf.gkg','rot_13')\n", "pt=codecs.decode('fep/pbasvtf','rot_13')\n", "#@markdown This cell can take up to 5 minutes to finish\n", - "!pip install pip==23.3.1\n", + "!pip install uv\n", + "!uv venv --python 3.10.12 .venv\n", + "!uv pip install pip==23.3.1 --quiet\n", + "\n", + "!uv pip install setuptools wheel ipykernel --quiet\n", + "\n", + "!uv pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/ --quiet\n", + "\n", "if Pitch_Change==\"12\":\n", " print(\"Using Pitch 12\")\n", " %cd $pt\n", @@ -133,7 +141,7 @@ " !sed -i '/tqdm==/d' $req\n", "\n", " time.sleep(5)\n", - " !pip install -r requirements.txt\n", + " !uv pip install -r requirements.txt --quiet\n", " # Fix Numpy deprecation\n", " !sed -i 's/(np.int)/(int)/g' '../vc_infer_pipeline.py'\n", " %cd ../../\n", @@ -150,10 +158,10 @@ " !sed -i '/tqdm==/d' $req\n", "\n", " time.sleep(5)\n", - " !pip install -r requirements.txt\n", + " !uv pip install -r requirements.txt --quiet\n", " # Fix Numpy deprecation\n", " !sed -i 's/(np.int)/(int)/g' 'src/vc_infer_pipeline.py'\n", - "!pip install gradio==3.48.0\n", + "!uv pip install gradio==3.48.0 --quiet\n", "\n", "\n", "\n", @@ -191,7 +199,7 @@ "#@markdown - Click [this link](https://dashboard.ngrok.com/get-started/your-authtoken) to get your auth token, and place it here:\n", "Token=\"\" #@param {type:\"string\"}\n", "if Url==\"Gradio\":\n", - " !python $runpice --share\n", + " !source .venv/bin/activate; python $runpice --share\n", "elif Url == \"Cloudflared\":\n", " # Install Cloudflared\n", " !curl -LO https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64.deb\n", @@ -205,16 +213,16 @@ " time.sleep(5)\n", "\n", " !grep -oE \"https://[a-zA-Z0-9.-]+\\.trycloudflare\\.com\" nohup.out\n", - " !python $runpice --listen-port 9999\n", + " !source .venv/bin/activate; python $runpice --listen-port 9999\n", "else:\n", - " !pip install pyngrok\n", + " !uv pip install pyngrok\n", " from pyngrok import ngrok\n", " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", " cover_tunnel = ngrok.connect(9999)\n", " print(\"CoverGen URL:\", cover_tunnel.public_url)\n", " print(credit)\n", - " !python $runpice --listen-port 9999" + " !source .venv/bin/activate; python $runpice --listen-port 9999" ] }, { From 4b8eb8f9f931011d169709a20d7440cfd5c6ac6a Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 17 Jan 2025 23:40:09 +0800 Subject: [PATCH 47/48] HotFix cuda broken --- Hina_Mod_AICoverGen_colab.ipynb | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index 131e9a6e..bea30103 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -121,12 +121,13 @@ "#@markdown This cell can take up to 5 minutes to finish\n", "!pip install uv\n", "!uv venv --python 3.10.12 .venv\n", - "!uv pip install pip==23.3.1 --quiet\n", + "!uv pip install pip==23.3.1 > /dev/null 2>&1\n", "\n", - "!uv pip install setuptools wheel ipykernel --quiet\n", + "!uv pip install setuptools wheel ipykernel numpy onnxruntime-gpu onnxruntime > /dev/null 2>&1\n", "\n", - "!uv pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/ --quiet\n", + "!uv pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 > /dev/null 2>&1\n", "\n", + "clear_output()\n", "if Pitch_Change==\"12\":\n", " print(\"Using Pitch 12\")\n", " %cd $pt\n", @@ -141,7 +142,9 @@ " !sed -i '/tqdm==/d' $req\n", "\n", " time.sleep(5)\n", - " !uv pip install -r requirements.txt --quiet\n", + " !uv pip install -r requirements.txt > /dev/null 2>&1\n", + " clear_output()\n", + "\n", " # Fix Numpy deprecation\n", " !sed -i 's/(np.int)/(int)/g' '../vc_infer_pipeline.py'\n", " %cd ../../\n", @@ -158,15 +161,17 @@ " !sed -i '/tqdm==/d' $req\n", "\n", " time.sleep(5)\n", - " !uv pip install -r requirements.txt --quiet\n", + " !uv pip install -r requirements.txt > /dev/null 2>&1\n", + " clear_output()\n", + "\n", " # Fix Numpy deprecation\n", " !sed -i 's/(np.int)/(int)/g' 'src/vc_infer_pipeline.py'\n", - "!uv pip install gradio==3.48.0 --quiet\n", + "!uv pip install gradio==3.48.0 > /dev/null 2>&1\n", "\n", "\n", "\n", "# install cuda fix\n", - "!python -m pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/\n", + "!uv pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/ > /dev/null 2>&1\n", "# clear_output()\n", "print(\"Finished Installing Requirements\")\n", "!sudo apt update\n", @@ -215,7 +220,7 @@ " !grep -oE \"https://[a-zA-Z0-9.-]+\\.trycloudflare\\.com\" nohup.out\n", " !source .venv/bin/activate; python $runpice --listen-port 9999\n", "else:\n", - " !uv pip install pyngrok\n", + " !uv pip install pyngrok > /dev/null 2>&1\n", " from pyngrok import ngrok\n", " ngrok.set_auth_token(Token)\n", " ngrok.kill()\n", From 26ddd219f9bd6b90b22e9fd1adae5bf7baec4742 Mon Sep 17 00:00:00 2001 From: Hina <79749008+hinabl@users.noreply.github.com> Date: Fri, 31 Jan 2025 20:52:22 +0800 Subject: [PATCH 48/48] Downgraded Torch --- Hina_Mod_AICoverGen_colab.ipynb | 1 + 1 file changed, 1 insertion(+) diff --git a/Hina_Mod_AICoverGen_colab.ipynb b/Hina_Mod_AICoverGen_colab.ipynb index bea30103..ca55e0e0 100644 --- a/Hina_Mod_AICoverGen_colab.ipynb +++ b/Hina_Mod_AICoverGen_colab.ipynb @@ -172,6 +172,7 @@ "\n", "# install cuda fix\n", "!uv pip install ort-nightly-gpu --index-url=https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ort-cuda-12-nightly/pypi/simple/ > /dev/null 2>&1\n", + "!uv pip install torch==2.0.1 > /dev/null 2>&1\n", "# clear_output()\n", "print(\"Finished Installing Requirements\")\n", "!sudo apt update\n",