search

NavodPeiris / speechlib

speechlib is a library that can do speaker diarization, transcription and speaker recognition on an audio file to create transcripts with actual speaker names
MIT License
114 stars 10 forks source link

Dev #39

Closed NavodPeiris closed 1 month ago

NavodPeiris commented 1 month ago
  1. added fp16 support when running regular whisper on gpu
  2. updated readme