fix tokenizer is None

2025-07-01 18:17:27 -04:00 · 2023-05-20 15:52:45 +02:00
parent 715435db42
commit a1c705b3a7
1 changed files with 7 additions and 4 deletions
--- a/whisperx/asr.py
+++ b/whisperx/asr.py
@ -231,6 +231,9 @@ class FasterWhisperPipeline(Pipeline):
        if self.tokenizer is None:
            language = language or self.detect_language(audio)
            task = task or "transcribe"
+            self.tokenizer = faster_whisper.tokenizer.Tokenizer(self.model.hf_tokenizer,
+                                                                self.model.model.is_multilingual, task=task,
+                                                                language=language)
        else:
            language = language or self.tokenizer.language_code
            task = task or self.tokenizer.task