schsu/breeze-asr-25-mlx

Breeze-ASR-25 MLX

This is an MLX-converted version of MediaTek-Research/Breeze-ASR-25, optimized for inference on Apple Silicon Macs using the MLX framework.

No PyTorch or model conversion required — download and run directly with mlx-whisper.

Model Details

Property	Value
Base model	MediaTek-Research/Breeze-ASR-25
Architecture	Whisper Large (32 encoder + 32 decoder layers)
Parameters	~1.5B
Format	MLX safetensors (float16)
Size	~2.9 GB
Language	Traditional Chinese (Taiwan Mandarin)
License	Apache 2.0

Usage

With mlx-whisper

import mlx_whisper

result = mlx_whisper.transcribe(
    "audio.wav",
    path_or_hf_repo="schsu/breeze-asr-25-mlx",
    language="zh",
)
print(result["text"])

With TypelessMLX App

This model is the default high-accuracy option in TypelessMLX, a macOS dictation app powered by MLX Whisper.

Performance

• Optimized for Taiwan Mandarin (繁體中文 / 台灣國語)
• Runs fully on-device on Apple Silicon (M1 and later)
• Uses fp16 precision for faster inference on Apple Neural Engine
• Significantly better accuracy on Traditional Chinese than standard Whisper Large v3

Conversion

Converted from the original PyTorch weights using weight remapping (HuggingFace Transformers → OpenAI Whisper naming convention) and tensor transposition (Conv1d channels).

Original model: MediaTek-Research/Breeze-ASR-25

Citation

If you use this model, please cite the original Breeze-ASR-25:

@misc{breeze-asr-25,
  author = {MediaTek Research},
  title  = {Breeze-ASR-25},
  year   = {2025},
  url    = {https://huggingface.co/MediaTek-Research/Breeze-ASR-25}
}