Jarod/ai-voice-cloning
1
1
Fork 0
forked from mrq/ai-voice-cloning
Code Pull Requests Activity

now whisperx should output json that aligns with what's expected

Browse Source
This commit is contained in:
mrq 2023-03-22 20:01:30 +00:00
parent 8877960062
commit 13605f980c
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/utils.py
View File

@ -1150,6 +1150,7 @@ def whisper_transcribe( file, language=None ):
segments = whisper_model.extract_text_and_timestamps( res ) segments = whisper_model.extract_text_and_timestamps( res )
result = { result = {
'text': []
'segments': [] 'segments': []
} }
for segment in segments: for segment in segments:
@ -1157,8 +1158,12 @@ def whisper_transcribe( file, language=None ):
'start': segment[0] / 100.0, 'start': segment[0] / 100.0,
'end': segment[1] / 100.0, 'end': segment[1] / 100.0,
'text': segment[2], 'text': segment[2],
'id': len(result['segments'])
} }
result['text'].append( segment[2] )
result['segments'].append(reparsed) result['segments'].append(reparsed)
result['text'] = " ".join(result['text'])
return result return result
if args.whisper_backend == "m-bain/whisperx": if args.whisper_backend == "m-bain/whisperx":
@ -1194,6 +1199,7 @@ def whisper_transcribe( file, language=None ):
result['segments'] = result_aligned['segments'] result['segments'] = result_aligned['segments']
result['text'] = [] result['text'] = []
for segment in result['segments']: for segment in result['segments']:
segment['id'] = len(result['text'])
result['text'].append(segment['text'].strip()) result['text'].append(segment['text'].strip())
result['text'] = " ".join(result['text']) result['text'] = " ".join(result['text'])