README.md

---
library_name: mlx-audio-plus
base_model:
- ResembleAI/chatterbox-turbo
tags:
- mlx
- tts
- chatterbox
pipeline_tag: text-to-speech
language:
- en
---

# Chatterbox-Turbo-TTS-4bit

This model was converted to MLX format from [ResembleAI/chatterbox-turbo](https://huggingface.co/ResembleAI/chatterbox-turbo) using [mlx-audio-plus](https://github.com/DePasqualeOrg/mlx-audio-plus) version **0.1.6**.

This model uses **4-bit quantization** for the T3 GPT2 backbone, reducing memory usage while maintaining audio quality.

**Note:** This model requires the S3Tokenizer weights from [mlx-community/S3TokenizerV2](https://huggingface.co/mlx-community/S3TokenizerV2), which will be downloaded automatically.

## Use with mlx-audio-plus

```bash
pip install -U mlx-audio-plus
```

### Command line

```bash
mlx_audio.tts --model /path/to/Chatterbox-Turbo-TTS-4bit --text "Hello, this is Chatterbox Turbo on MLX!" --ref_audio reference.wav
```

### Python

```python
from mlx_audio.tts.generate import generate_audio

generate_audio(
    text="Hello, this is Chatterbox Turbo on MLX!",
    model="/path/to/Chatterbox-Turbo-TTS-4bit",
    ref_audio="reference.wav",
    file_prefix="output",
)
```