Skip to content

fxfabre/doc_to_speech

Repository files navigation

doc_to_speech

Book reader : From word / pdf to audio

Un dossier par fonctionnalité :

  • ocr : reconnaissance de caractères sur un pdf / une image
  • text_to_speech : TTS avec suno / bark

Text to speech

Sources :

Configure Accelerate

  • To optimize GPU usage : accelerate
  • Run accelerate config & check config : accelerate env
  • Config file at ./model/accelerate/default_config.yaml

Setup project

  1. Create .env file with :
    SUNO_USE_SMALL_MODELS=true
    SUNO_ENABLE_MPS=true
    HF_HOME=./model_cache
    
  2. Install uv
  3. Install dependencies : uv pip install -r requirements.txt
  4. Run : python main.py

About

From word / pdf to audio

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages