This script takes an epub (or text file) and reads it to an m4b audiobook file, using TTS by https://github.com/coqui-ai/TTS or OpenAI. The audiofiles are created in discrete chunks then transcribed using whisper speech-to-text. The transcription is compared to the original text, and if they don't match well it tries again. Finally all silence longer than a second is removed from all audio segments, and the audio is cleaned up before being combined into an m4b audiobook file.
I recognize this is not very user friendly, but I wanted to share in case folks thought it was useful. If there are a few more people than myself that find this is useful I will keep working on turning it into something that could be used by someone without dev experience.
NOTE: Latest release adds a new workflow allowing you to export the epub to text, make any necessary modifications, then read the book as a text file. Any line beginning with "# " is considered a chapter break, and will be automatically inserted during export, named "# Part 1", etc. If you replace "Part 1" with whatever you want that section to be called it will be labeled that way in the audiobook metadata.
NOTE: DeepSpeed support for XTTS has been added! If deepspeed is installed and you have a compatible GPU, it will be detected and used. For XTTS, this will yeild a 3x-4x speed improvement! Install deepspeed with pip install deepspeed.
NOTE: The Coqui team released their curated XTTS voice models recently, and they sound great. A recent update here
allows you to use these voices. You can generate samples of all the voices by running python utils/generate-speaker-samples.py. Check these voices out, they're allmost all amazing sounding! (GPU required). Also samples of the available XTTS voices, without installing the package first can be found there:
https://github.com/rejuce/CoquiTTS_XTTS_Examples
Example usage: epub2tts my-book.epub --engine xtts --speaker "Damien Black" --cover cover-image.jpg
NOTE: The Coqui team released v2 of their XTTS model and the quality is amazing! This latest release includes significant refactoring, and uses streaming inference for XTTS. Suggested usage is to include up to three wav file speaker samples, up to 30 seconds each. Check out the XTTS sample to get an idea of the quality you can expect. Also take a look in the utils directory for notes on finetuning your model for exceptional results. (GPU required)
Example usage: epub2tts my-book.epub --start 4 --end 20 --xtts shadow-1.wav,shadow-2.wav,shadow-3.wav --cover cover-image.jpg
Usage:
EPUB: epub2tts my-book.epub --cover cover-image.jpg
EXPORT: epub2tts my-book.epub --export txt
TEXT: epub2tts my-book.txt
To use Coqui XTTS, add: --xtts <sample-1.wav>,<sample-2.wav>,<sample-3.wav> --language 'en' book.epub (slow but sounds amazing!)
To use OpenAI TTS, add: --openai <your API key> (Use speaker option to specify voice other than onyx: --speaker shimmer)
To change speaker (ex p307 for a good male voice w/Coqui TTS), add: --speaker p307
To skip reading any links, add: --skiplinks
Using --scan will list excerpts of each chapter, then exit. This is helpful for finding which chapter to start and end on if you want to skip bibliography, TOC, etc.
Using --export txt will export the entire book to text file. This will honor --start and --end arguments as well.
To specify which chapter to start on (ex 3): --start 3
To specify which chapter to end on (ex 20): --end 20
To specify bitrate (ex 30k): --bitrate 30k
To specify minimum comparison ratio between transcript of spoken text and original, default 88. Set to 0 to disable this comparison with whisper: --minratio 95
To embed a cover image with the audiobook, add: --cover your-cover.jpg
If epub2tts is interrupted or crashes, you can run it again with the same parameters and it will pick up where it left off, assuming it made it far enough to save some WAV files. If you want to start fresh, be sure to delete any of the wav files (with the same name as the epub) in the working directory before running again.
Voice models will be saved locally in ~/.local/share/tts
For Linux and MacOS:
alias epub2tts='docker run -v "$PWD:$PWD" -v ~/.local/share/tts:/root/.local/share/tts -w "$PWD" ghcr.io/aedocw/epub2tts:release'
For Windows: Pre-requisites:
- Install Docker Desktop
- From PowerShell run "mkdir ~/.local/share/tts"
#Example for running scan of "mybook.epub"
docker run -v ${PWD}/.local/share/tts:/root/.local/share/tts -v ${PWD}:/root -w /root ghcr.io/aedocw/epub2tts:release mybook.epub --scan
#Example for reading parts 3 through 15 of "mybook.epub"
docker run -v ${PWD}/.local/share/tts:/root/.local/share/tts -v ${PWD}:/root -w /root ghcr.io/aedocw/epub2tts:release mybook.epub --start 3 --end 15
This installation requires Python < 3.12 and Homebrew (I use homebrew to install espeak, pyenv and ffmpeg). Per this bug, mecab should also be installed via homebrew.
Voice models will be saved locally in ~/.local/share/tts
#install dependencies
brew install espeak pyenv ffmpeg mecab
#install epub2tts
git clone https://github.com/aedocw/epub2tts
cd epub2tts
pyenv install 3.11
pyenv local 3.11
#OPTIONAL - install this in a virtual environment
python -m venv .venv && source .venv/bin/activate
pip install .
These instructions are for Ubuntu, but should work (with appropriate package installer mods) for just about any repo. Ensure you have ffmpeg installed before use.
Voice models will be saved locally in ~/.local/share/tts
#install dependencies
sudo apt install espeak-ng ffmpeg
#clone the repo
git clone https://github.com/aedocw/epub2tts
cd epub2tts
pip install .
#clone the repo
git clone https://github.com/aedocw/epub2tts
cd epub2tts
#create a virtual environment
python -m venv .venv
#activate the virtual environment
source .venv/bin/activate
#install dependencies
sudo apt install espeak-ng ffmpeg
pip install -r requirements.txt