speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

update pitch

Browse Source
This commit is contained in:
维石 2024-09-30 17:21:00 +08:00
parent 1ca3c42282
commit 7fcca8f23a
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
funasr/datasets/sense_voice_datasets/datasets.py
View File

@ -587,10 +587,10 @@ class SenseVoicePitchDataset(torch.utils.data.Dataset):
elif f0.shape < speech.shape[1]:
last_value = f0[-1]
f0 = torch.cat([f0, last_value.repeat(speech.shape[1] - f0.shape)])
f0_tag = torch.Tensor([1], dtype=torch.int32)
f0_tag = torch.tensor([1], dtype=torch.int32)
else:
f0 = torch.tensor([0.0], dtype=torch.float32)
f0_tag = torch.Tensor([0], dtype=torch.int32)
f0_tag = torch.tensor([0], dtype=torch.int32)
output = {
"speech": speech[0, :, :],