DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism

This repository is the official PyTorch implementation of our AAAI-2022 paper, in which we propose DiffSinger (for Singing-Voice-Synthesis) and DiffSpeech (for Text-to-Speech).

🎉 🎉 🎉 Updates:

• Sep.11, 2022: 🔌 DiffSinger-PN. Add plug-in PNDM, ICLR 2022 in our laboratory, to accelerate DiffSinger freely.
• Jul.27, 2022: Update documents for SVS. Add easy inference A & B; Add Interactive SVS running on HuggingFace🤗 SVS.
• Mar.2, 2022: MIDI-B-version.
• Mar.1, 2022: NeuralSVB, for singing voice beautifying, has been released.
• Feb.13, 2022: NATSpeech, the improved code framework, which contains the implementations of DiffSpeech and our NeurIPS-2021 work PortaSpeech has been released.
• Jan.29, 2022: support MIDI-A-version SVS.
• Jan.13, 2022: support SVS, release PopCS dataset.
• Dec.19, 2021: support TTS. HuggingFace🤗 TTS

🚀 News:

• Feb.24, 2022: Our new work, NeuralSVB was accepted by ACL-2022 . Demo Page.
• Dec.01, 2021: DiffSinger was accepted by AAAI-2022.
• Sep.29, 2021: Our recent work PortaSpeech: Portable and High-Quality Generative Text-to-Speech was accepted by NeurIPS-2021 .
• May.06, 2021: We submitted DiffSinger to Arxiv .

Environments

1. If you want to use env of anaconda:

   conda create -n your_env_name python=3.8
   source activate your_env_name 
   pip install -r requirements_2080.txt   (GPU 2080Ti, CUDA 10.2)
   or pip install -r requirements_3090.txt   (GPU 3090, CUDA 11.4)

2. Or, if you want to use virtual env of python:

   ## Install Python 3.8 first. 
   python -m venv venv
   source venv/bin/activate
   # install requirements.
   pip install -U pip
   pip install Cython numpy==1.19.1
   pip install torch==1.9.0
   pip install -r requirements.txt

Documents

• Run DiffSpeech (TTS version).
• Run DiffSinger (SVS version).

Overview

Mel Pipeline	Dataset	Pitch Input	F0 Prediction	Acceleration Method	Vocoder
DiffSpeech (Text->F0, Text+F0->Mel, Mel->Wav)	Ljspeech	None	Explicit	Shallow Diffusion	HiFiGAN
DiffSinger (Lyric+F0->Mel, Mel->Wav)	PopCS	Ground-Truth F0	None	Shallow Diffusion	NSF-HiFiGAN
DiffSinger (Lyric+MIDI->F0, Lyric+F0->Mel, Mel->Wav)	OpenCpop	MIDI	Explicit	Shallow Diffusion	NSF-HiFiGAN
FFT-Singer (Lyric+MIDI->F0, Lyric+F0->Mel, Mel->Wav)	OpenCpop	MIDI	Explicit	Invalid	NSF-HiFiGAN
DiffSinger (Lyric+MIDI->Mel, Mel->Wav)	OpenCpop	MIDI	Implicit	None	Pitch-Extractor + NSF-HiFiGAN
DiffSinger+PNDM (Lyric+MIDI->Mel, Mel->Wav)	OpenCpop	MIDI	Implicit	PLMS	Pitch-Extractor + NSF-HiFiGAN
DiffSpeech+PNDM (Text->Mel, Mel->Wav)	Ljspeech	None	Implicit	PLMS	HiFiGAN

Tensorboard

tensorboard --logdir_spec exp_name

Citation

@article{liu2021diffsinger,
  title={Diffsinger: Singing voice synthesis via shallow diffusion mechanism},
  author={Liu, Jinglin and Li, Chengxi and Ren, Yi and Chen, Feiyang and Liu, Peng and Zhao, Zhou},
  journal={arXiv preprint arXiv:2105.02446},
  volume={2},
  year={2021}}

Acknowledgements

• lucidrains' denoising-diffusion-pytorch
• Official PyTorch Lightning
• kan-bayashi's ParallelWaveGAN
• jik876's HifiGAN
• Official espnet
• lmnt-com's DiffWave
• keonlee9420's Implementation.

Especially thanks to:

• Team Openvpi's maintenance: DiffSinger.
• Your re-creation and sharing.