Martin Källström
Results
2
comments of
Martin Källström
Also looking forward to this!
Do you use the large-v3 tokenizer as described by https://huggingface.co/turicas/faster-whisper-large-v3 ? Perhaps run your test using this model as well.