Autiobooks generates .m4b audiobooks from regular .epub e-books, using Kokoro's high-quality speech synthesis.
Kokoro is an open-weight text-to-speech model with 82 million parameters. It yields natural sounding output while being able to run on consumer hardware.
It supports American, British English, French, Korean, Japanese and Mandarin (though we only-support English, for now) and a wide range of different voices with different accents and prosody.
PRs are welcome!
If you have Python 3 on your computer, you can install it with pip. Be aware that it won't work with Python 3.13.
pip install autiobooksYou will require ffmpeg and tkinter installed:
Linux:
sudo apt install ffmpeg python3-tkinterMacOS:
brew install ffmpeg python3-tkinterTo start the program, run:
autiobooksThe program creates .wav files for each chapter, then combines them into a .m4b file for playing using an audiobook player.
by David Nesbitt, distributed under MIT license. Check out the excellent project audiblez if you'd prefer a command-line interface.
