Skip to content

Speech recognition with word-level timestamps, optimized for batch inference.

Notifications You must be signed in to change notification settings

kb-labb/easytranscriber

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Easywhisper

Easywhisper is a transcription library which decomposes the inference pipeline into independent and parallelizable components (VAD, transcription, feature/emission extraction, forced alignment). The library is therefore well suited for transcribing large archives of audio files efficiently. Supports both ctranslate2 (faster-whisper) and Hugging Face backends for inference. The library features:

  • Batch inference support for both wav2vec2 and Whisper models.
  • Parallel loading of audio files for efficient batch processing.
  • GPU accelerated forced alignment.
  • Saves wav2vec2 emmissions to disk for flexible parallel processing on CPU/GPU.

About

Speech recognition with word-level timestamps, optimized for batch inference.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages