Mirroristas/whisper
2
0
Fork 0
mirror of https://github.com/openai/whisper.git synced 2025-11-24 06:26:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

Avoid computing higher temperatures on no_speech segments (#1279)

Browse Source 
* Avoid computing higher temperatures on no_speech

In decode_with_fallback, we compute higher temperatures in the case where compression_ratio is too high or avg_logprob is too low.
But as the computation of no_speech_prob doens't depend on sampling, we can avoid computing higher temperatures if we detect in the first one that the no_speech condition is fulfilled

* Update transcribe.py

---------

Co-authored-by: Jong Wook Kim <jongwook@openai.com>
This commit is contained in:
Théo BOYER 2023-05-05 02:02:36 +02:00 committed by GitHub
parent 5523722842
commit e334ff141d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
whisper/transcribe.py
View File

@ -174,7 +174,11 @@ def transcribe(
and decode_result.avg_logprob < logprob_threshold and decode_result.avg_logprob < logprob_threshold
): ):
needs_fallback = True # average log probability is too low needs_fallback = True # average log probability is too low
if (
no_speech_threshold is not None
and decode_result.no_speech_prob > no_speech_threshold
):
needs_fallback = False # silence
if not needs_fallback: if not needs_fallback:
break break