Text-to-Speech
Safetensors
MLX
mlx-audio
moonbox
fish-audio
fish-speech
s2-pro
voice-cloning
multilingual
Instructions to use moonbox-dev/s2-pro-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use moonbox-dev/s2-pro-MLX with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download moonbox-dev/s2-pro-MLX --local-dir s2-pro-MLX
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
|
Download NOTICE.md from moonbox-dev/s2-pro-MLX: direct link, hf CLI and curl.
- Browser
- Download file 522 Bytes
-
https://huggingface.co/moonbox-dev/s2-pro-MLX/resolve/main/NOTICE.md
- Command line
-
hf download hf://moonbox-dev/s2-pro-MLX/NOTICE.md
-
curl -L -o NOTICE.md https://huggingface.co/moonbox-dev/s2-pro-MLX/resolve/main/NOTICE.md
522 Bytes
This model is licensed under the Fish Audio Research License, Copyright © 39 AI, INC. All Rights Reserved.
Built with Fish Audio.
Changes by Moonbox (moonbox-dev): the weights of fishaudio/s2-pro (revision 1de9996b6be38b745688de084d87a5633f714e4e) were converted to MLX for mlx-audio: stored as bf16 or fp16, or with the transformer layers quantized to 8, 6 or 4 bits (MLX affine, group size 64); the codec (codec.pth) was re-saved as codec.safetensors (float32, unchanged values). Nothing was retrained or fine-tuned.