---
title: VibeVoice-ASR-5bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/VibeVoice-ASR-5bit"
md_url: "https://www.modelscope.cn/models/mlx-community/VibeVoice-ASR-5bit.md"
repository: mlx-community/VibeVoice-ASR-5bit
last_updated: 2026-01-23
license: mit
model_type:
  - vibevoice_asr
architectures:
  - VibeVoiceForASRTraining
parameters: 358.9M
tensor_type:
  - U32
  - BF16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 76
stars: 0
tags:
  - ASR
  - Transcriptoin
  - Diarization
  - Speech-to-Text
  - mlx
  - speech-to-text
  - speech
  - transcription
  - asr
  - stt
  - mlx-audio
---

# VibeVoice-ASR-5bit

> VibeVoice-ASR-5bit - mlx-community 在 ModelScope 开源的模型。mlx-community/VibeVoice-ASR-5bit

mlx-community/VibeVoice-ASR-5bit 是 ModelScope 魔搭社区上的 358.9M 参数机器学习模型，采用 mit 许可。

- **Repository**: mlx-community/VibeVoice-ASR-5bit
- **License**: mit
- **Parameters**: 358.9M
- **Tags**: ASR, Transcriptoin, Diarization, Speech-to-Text, mlx, speech-to-text, speech, transcription, asr, stt, mlx-audio
- **Downloads**: 76
- **Stars**: 0
- **Last updated**: 2026-01-23

Source: https://www.modelscope.cn/models/mlx-community/VibeVoice-ASR-5bit

---

# mlx-community/VibeVoice-ASR-5bit

This model was converted to MLX format from [`microsoft/VibeVoice-ASR`](https://huggingface.co/microsoft/VibeVoice-ASR) using mlx-audio version **0.3.0**.

Refer to the [original model card](https://huggingface.co/microsoft/VibeVoice-ASR) for more details on the model.

## Use with mlx-audio

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

### CLI Example:
```bash
python -m mlx_audio.stt.generate --model mlx-community/VibeVoice-ASR-5bit --audio "audio.wav"
```

### Python Example:
```python        
from mlx_audio.stt.utils import load_model
from mlx_audio.stt.generate import generate_transcription

model = load_model("mlx-community/VibeVoice-ASR-5bit")
transcription = generate_transcription(
    model=model,
    audio_path="path_to_audio.wav",
    output_path="path_to_output.txt",
    format="txt",
    verbose=True,
)
print(transcription.text)

```