Spaces:

Duplicated from  whisper-event/whisper-demo

vumichien

/

Whisper_speaker_diarization

like 419

Runtime error

Really like the project. However, getting following error when trying to run locally - what might I be doing wrong?

╭──────────────────────────── Traceback (most recent call last) ─────────────────────────────╮
│ /Users/Alfie/PycharmProjects/osteonotes/Whisper_speaker_diarization/app.py:144 in │
│ │
│ 141 │ device=device, │
│ 142 ) │
│ 143 os.makedirs('output', exist_ok=True) │
│ ❱ 144 pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(languag │
│ 145 │
│ 146 embedding_model = PretrainedSpeakerEmbedding( │
│ 147 │ "speechbrain/spkrec-ecapa-voxceleb", │
╰────────────────────────────────────────────────────────────────────────────────────────────╯
AttributeError: 'WhisperTokenizer' object has no attribute 'get_decoder_prompt_ids'