mirrors/whispering
1
0
Fork 0
mirror of https://github.com/shirayu/whispering.git synced 2025-02-16 10:35:16 +00:00
Code Issues Projects Releases Packages Wiki Activity

Fix the behavior without --allow-padding

Browse source
This commit is contained in:
Yuta Hayashibe 2022-10-02 22:46:01 +09:00
parent 9fb789695e
commit 3221153415
1 changed files with 11 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
whispering/transcriber.py
View file

@ -260,13 +260,22 @@ class WhisperStreamingTranscriber:
seek: int = 0
while seek < mel.shape[-1]:
logger.debug(f"seek: {seek}")
if mel.shape[-1] - seek < N_FRAMES:
logger.debug(
f"mel.shape ({mel.shape[-1]}) - seek ({seek}) < N_FRAMES ({N_FRAMES})"
)
if ctx.allow_padding:
logger.warning("Padding is not expected while speaking")
else:
logger.debug("No padding")
break
segment: torch.Tensor = (
pad_or_trim(mel[:, seek:], N_FRAMES)
.to(self.model.device) # type: ignore
.to(self.dtype)
)
if not ctx.allow_padding and segment.shape[-1] > mel.shape[-1]:
logger.warning("Padding is not expected while speaking")
logger.debug(
f"seek={seek}, timestamp={ctx.timestamp}, "
@ -309,9 +318,6 @@ class WhisperStreamingTranscriber:
seek += last_timestamp_position * self.input_stride
logger.debug(f"new seek={seek}, mel.shape: {mel.shape}")
if (not ctx.allow_padding) and (mel.shape[-1] - seek < N_FRAMES):
break
if mel.shape[-1] - seek <= 0:
logger.debug(f"ctx.buffer_mel is None ({mel.shape}, {seek})")
return