Speechlib Versions Save

speechlib is a library that can do speaker diarization, transcription and speaker recognition on an audio file to create transcripts with actual speaker names

1.0.15

2 weeks ago

using faster-whisper==1.0.2

1.0.14

3 weeks ago

Adds MPS support for faster inferencing on Macs

1.0.13

3 months ago

1.0.11

4 months ago

changelog:

log file saved as filename_HHMMSS_language.txt

1.0.10

4 months ago

change log:

  • added faster-whisper
  • now able to directly input mp3, flac files to transcribe function without converting to wav
  • updated readme

1.0.7

4 months ago

changed readme and library.md

1.0.6

4 months ago

added examples

1.0.5

4 months ago

changed a bit

1.0.4

4 months ago

added whisper tiny, large model support added preprocessor functions changed readme

1.0.3

4 months ago

fixed issues with files having different encoding and multiple channels.