# Qwen3 Audiobook Converter

Convert TXT, PDF, EPUB, DOCX, and DOC files into audiobooks using the Qwen3-TTS voice model.

Original project: [https://github.com/WhiskeyCoder/Qwen3-Audiobook-Converter](https://github.com/WhiskeyCoder/Qwen3-Audiobook-Converter
). This repo just has minor fixes, flags, and documentation updates. It also splits the qwen3-tts server into two processes running models on different ports.

## Overview

The converter sends text extracted from your books to a locally running Qwen3-TTS server and assembles the returned audio into a single audiobook file.

- Supported input: `.txt`, `.pdf`, `.epub`, `.docx`, `.doc`
- Output: `.mp3`
- Two voice modes:
  - Custom voice: pre-built speakers
  - Voice clone: clone a voice from a `.wav` reference audio file

## Prerequisites

- Python 3.12
- ffmpeg
- Enough VRAM to run the 1.7B model (~6GB)

## Installation

```bash
# Arch Linux
sudo pacman -S conda ffmpeg

# Debian, conda must be installed separately
sudo apt-get install ffmpeg
```

### Install Qwen3-TTS (Server)

```bash
conda create -n qwen3-tts python=3.12 -y
conda activate qwen3-tts
pip install -U qwen-tts
```

### Install the conversion script

```bash
git clone https://git.historia.vg/git/qwen3-audiobook-converter
cd qwen3-audiobook-converter
pip install -r requirements.txt
```

## Running the Qwen-TTS server

The converter script talks to a Qwen3-TTS Gradio server that is run using `qwen-tts-demo`. Add `--no-flash-attn` if FlashAttention isn't installed (see below). The script expects the custom voice model and base model to be on different ports depending on which you're using:

### Custom voice

```bash
conda activate qwen3-tts
qwen-tts-demo Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice --ip 127.0.0.1 --port 7860
```

### Voice clone

```bash
conda activate qwen3-tts
qwen-tts-demo Qwen/Qwen3-TTS-12Hz-1.7B-Base --ip 127.0.0.1 --port 7861
```

## Converting books

Put your book files (epub, txt, etc.) in the `book_to_convert/` folder. Then run the script. The output mp3 goes to `audiobooks/`.

### Custom voice

```bash
python audiobook_converter.py
```

Edit the parameters at the top of `audiobook_converter.py` to change which built-in voice is used.

```
CUSTOM_VOICE_SPEAKER = "Vivian"   # Serena, Vivian, Uncle_Fu, Aiden, Ono_Anna, Sohee, Eric, Dylan
CUSTOM_VOICE_LANGUAGE = "English"
CUSTOM_VOICE_INSTRUCT = "Speak naturally and clearly, as if reading a dramatic book to an adult audience."
```

### Voice clone

```bash
python audiobook_converter.py --voice-clone --voice-sample path/to/reference.wav
```

The reference .wav should be ~10-15 seconds with a minimum of 3 seconds and maximum of 60 seconds. Longer is not better. ~15 seconds is ideal.

Whisper will be used automatically to transcribe the reference audio (`faster_whisper` or `whisper`). If no Whisper backend is installed, it falls back to x-vector-only cloning.

To skip automatic transcription explicitly, pass `--no-transcription`. This should be worse, but in my experience may give a preferable flatter tone to certain voices.

You can override whisper by passing your own transcription with `--voice-sample-text "What the reference audio says"`

### Playback speed

Adjust the speed of the final audiobook without changing pitch (uses ffmpeg `atempo` before encoding). The normal-speed audiobook file is also preserved in the output directory.

```bash
python audiobook_converter.py --speed 0.9
```

## FlashAttention (optional)

The server tries to use FlashAttention 2 by default, but `--no-flash-attn` works without it. On supported GPUs FlashAttention can give a modest speedup.

1. Build from source (takes absolutely forever). If you run out of memory, lower MAX_JOBS until you don't.

```bash
conda activate qwen3-tts
pip install ninja packaging psutil
MAX_JOBS=4 pip install --no-build-isolation flash-attn
```

2. Or pip install a prebuilt wheel matching your torch / CUDA / Python / CXX11-ABI combination:

```bash
python -c "import torch; print(torch.__version__, torch.version.cuda, torch._C._GLIBCXX_USE_CXX11_ABI)"
```

Official wheels: https://github.com/Dao-AILab/flash-attention/releases (pick `cp312` + matching `cuX` + `torchX.Y` + `cxx11abiTRUE/FALSE`).

Third-party wheels: https://mjunya.com/flash-attention-prebuild-wheels/ (hosted at https://github.com/mjun0812/flash-attention-prebuild-wheels).

## License

MIT

## Credits

- [Qwen3-Audiobook-Converter](https://github.com/WhiskeyCoder/Qwen3-Audiobook-Converter) by WhiskeyCoder.
- [Qwen3-TTS](https://github.com/QwenLM/Qwen3-TTS) voice model.