chrissewell/whisperX
1
0
Fork 0
mirror of https://github.com/m-bain/whisperX.git synced 2025-07-01 18:17:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

add tokenizer is None case

Browse Source
This commit is contained in:
Simon
2023-05-20 15:42:21 +02:00
parent 1fc965bc1a
commit 715435db42
1 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
whisperx/asr.py
View File

@ -228,9 +228,12 @@ class FasterWhisperPipeline(Pipeline):
vad_segments = self.vad_model({"waveform": torch.from_numpy(audio).unsqueeze(0), "sample_rate": SAMPLE_RATE}) vad_segments = self.vad_model({"waveform": torch.from_numpy(audio).unsqueeze(0), "sample_rate": SAMPLE_RATE})
vad_segments = merge_chunks(vad_segments, 30) vad_segments = merge_chunks(vad_segments, 30)
if self.tokenizer is None:
language = language or self.tokenizer.language_code language = language or self.detect_language(audio)
task = task or self.tokenizer.task task = task or "transcribe"
else:
language = language or self.tokenizer.language_code
task = task or self.tokenizer.task
if task != self.tokenizer.task or language != self.tokenizer.language_code: if task != self.tokenizer.task or language != self.tokenizer.language_code:
self.tokenizer = faster_whisper.tokenizer.Tokenizer(self.model.hf_tokenizer, self.tokenizer = faster_whisper.tokenizer.Tokenizer(self.model.hf_tokenizer,
self.model.model.is_multilingual, task=task, self.model.model.is_multilingual, task=task,