Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Using hamming window for Paraformer frontend. #2549

Merged
merged 1 commit into from
Jun 3, 2024
Merged

Using hamming window for Paraformer frontend. #2549

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion wenet/cli/paraformer_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,8 @@ def transcribe(self, audio_file: str, tokens_info: bool = False) -> dict:
frame_length=25,
frame_shift=10,
energy_floor=0.0,
sample_frequency=self.resample_rate)
sample_frequency=self.resample_rate,
window_type="hamming")
feats = feats.unsqueeze(0)
feats_lens = torch.tensor([feats.size(1)],
dtype=torch.int64,
Expand Down
6 changes: 4 additions & 2 deletions wenet/dataset/processor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -231,7 +231,8 @@ def compute_fbank(sample,
num_mel_bins=23,
frame_length=25,
frame_shift=10,
dither=0.0):
dither=0.0,
window_type="povey"):
""" Extract fbank

Args:
Expand All @@ -253,7 +254,8 @@ def compute_fbank(sample,
frame_shift=frame_shift,
dither=dither,
energy_floor=0.0,
sample_frequency=sample_rate)
sample_frequency=sample_rate,
window_type=window_type)
sample['feat'] = mat
return sample

Expand Down
1 change: 1 addition & 0 deletions wenet/paraformer/convert_paraformer_to_wenet_config_and_ckpt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,6 +140,7 @@ def convert_to_wenet_yaml(configs, wenet_yaml_path: str,
configs['dataset_conf']['fbank_conf']['frame_shift'] = 10
configs['dataset_conf']['fbank_conf']['frame_length'] = 25
configs['dataset_conf']['fbank_conf']['dither'] = 0.1
configs['dataset_conf']['fbank_conf']['window_type'] = 'hamming'
configs['dataset_conf']['spec_sub'] = False
configs['dataset_conf']['spec_trim'] = False
configs['dataset_conf']['shuffle'] = True
Expand Down
Loading