# ebook2audiobook
ebook2audiobook is an open-source tool that converts e-books into audiobooks using text-to-speech (TTS) engines. It supports voice cloning from custom audio samples and covers 1,158+ languages and dialects. Intended for use with non-DRM, legally acquired e-books only.
## Key Features
- **Wide format support**: Converts `.epub`, `.mobi`, `.azw3`, `.pdf`, `.txt`, and more
- **Multiple output formats**: `.m4b`, `.m4a`, `.mp3`, `.flac`, `.wav`, `.ogg`, `.aac`
- **Voice cloning**: Synthesize a custom voice from a provided audio sample
- **1,158+ languages**: Broad language and dialect coverage
- **OCR**: Can extract text from image-based pages
- **SML tags**: Fine-grained control with `[break]`, `[pause]`, and voice-switching tags
- **Low resource requirements**: 2GB RAM minimum, 8GB recommended
## TTS Engines
- **XTTSv2** (default for most languages)
- BARK
- VITS
- Fairseq
- Tacotron2
- YourTTS
## Installation
### Local
```bash
git clone https://github.com/DrewThomasson/ebook2audiobook.git
cd ebook2audiobook
# Linux/macOS
./ebook2audiobook.command
# Windows
ebook2audiobook.cmd
```
### Docker
Full containerized deployment with CPU, CUDA, and ROCm support.
### Cloud
Available on Hugging Face Spaces, Google Colab, and Kaggle.
## Tips
- Clean up your e-book before conversion — EPUB lacks a standard structure for chapters/paragraphs, so unwanted text (table of contents, copyright pages, etc.) may end up in the audio
- Voice cloning quality depends on the clarity of the input audio sample
## References
- GitHub: https://github.com/DrewThomasson/ebook2audiobook
## Related
- [[Text-to-Speech]]
- [[Audiobooks]]
- [[AI]]