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

vad bugfix (#624)

Browse Source 
Co-authored-by: 凌匀 <ailsa.zly@alibaba-inc.com>
This commit is contained in:
lingyunfly 2023-06-13 11:13:03 +08:00 committed by GitHub
parent 4a33f08def
commit 3da40ad9fe
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
funasr/models/e2e_vad.py
View File

@ -296,13 +296,14 @@ class E2EVadModel(nn.Module):
self.sil_frame = 0
self.frame_probs = []
assert self.output_data_buf[-1].contain_seg_end_point == True
drop_frames = int(self.output_data_buf[-1].end_ms / self.vad_opts.frame_in_ms)
real_drop_frames = drop_frames - self.last_drop_frames
self.last_drop_frames = drop_frames
self.data_buf_all = self.data_buf_all[real_drop_frames * int(self.vad_opts.frame_in_ms * self.vad_opts.sample_rate / 1000):]
self.decibel = self.decibel[real_drop_frames:]
self.scores = self.scores[:, real_drop_frames:, :]
if self.output_data_buf:
assert self.output_data_buf[-1].contain_seg_end_point == True
drop_frames = int(self.output_data_buf[-1].end_ms / self.vad_opts.frame_in_ms)
real_drop_frames = drop_frames - self.last_drop_frames
self.last_drop_frames = drop_frames
self.data_buf_all = self.data_buf_all[real_drop_frames * int(self.vad_opts.frame_in_ms * self.vad_opts.sample_rate / 1000):]
self.decibel = self.decibel[real_drop_frames:]
self.scores = self.scores[:, real_drop_frames:, :]
def ComputeDecibel(self) -> None:
frame_sample_length = int(self.vad_opts.frame_length_ms * self.vad_opts.sample_rate / 1000)