Text-to-Speech
Safetensors
MLX
mlx-audio
parler
annotation
speech
speech generation
voice cloning
tts
Instructions to use mlx-community/parler-tts-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/parler-tts-mini with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download mlx-community/parler-tts-mini --local-dir parler-tts-mini
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download preprocessor_config.json from mlx-community/parler-tts-mini: direct link, hf CLI and curl.
- Browser
- Download file 206 Bytes
-
https://huggingface.co/mlx-community/parler-tts-mini/resolve/main/preprocessor_config.json
- Command line
-
hf download hf://mlx-community/parler-tts-mini/preprocessor_config.json
-
curl -L -o preprocessor_config.json https://huggingface.co/mlx-community/parler-tts-mini/resolve/main/preprocessor_config.json
206 Bytes
| { | |
| "feature_extractor_type": "DacFeatureExtractor", | |
| "feature_size": 1, | |
| "hop_length": 512, | |
| "padding_side": "right", | |
| "padding_value": 0.0, | |
| "return_attention_mask": true, | |
| "sampling_rate": 44100 | |
| } | |