metadata
language:
- en
- zh
license: mit
pipeline_tag: automatic-speech-recognition
tags:
- ASR
- Transcriptoin
- Diarization
- Speech-to-Text
- mlx
- speech-to-text
- speech
- transcription
- asr
- stt
- mlx-audio
library_name: mlx-audio
mlx-community/VibeVoice-ASR-6bit
This model was converted to MLX format from microsoft/VibeVoice-ASR using mlx-audio version 0.3.0rc1.
Refer to the original model card for more details on the model.
Use with mlx-audio
pip install -U mlx-audio
CLI Example:
python -m mlx_audio.stt.generate --model mlx-community/VibeVoice-ASR-6bit --audio "audio.wav"
Python Example:
from mlx_audio.stt.utils import load_model
from mlx_audio.stt.generate import generate_transcription
model = load_model("mlx-community/VibeVoice-ASR-6bit")
transcription = generate_transcription(
model=model,
audio_path="path_to_audio.wav",
output_path="path_to_output.txt",
format="txt",
verbose=True,
)
print(transcription.text)