Automatic Speech Recognition
MLX
Safetensors
English
Chinese
mega_asr
speech-to-text
asr
robust-asr
qwen3-asr
quantized
int8
Instructions to use littlebearlabs/Mega-ASR-MLX-int8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use littlebearlabs/Mega-ASR-MLX-int8 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Mega-ASR-MLX-int8 littlebearlabs/Mega-ASR-MLX-int8
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
int8 affine-quantized Mega-ASR (group_size 64); router+LoRA dense; WER-parity gated
3a0151d verified | { | |
| "chunk_length": 30, | |
| "dither": 0.0, | |
| "feature_extractor_type": "WhisperFeatureExtractor", | |
| "feature_size": 128, | |
| "hop_length": 160, | |
| "n_fft": 400, | |
| "n_samples": 480000, | |
| "nb_max_frames": 3000, | |
| "padding_side": "right", | |
| "padding_value": 0.0, | |
| "processor_class": "Qwen3ASRProcessor", | |
| "return_attention_mask": true | |
| } |