metadata
base_model:
- Qwen/Qwen3-ASR-1.7B
datasets:
- knoveleng/cv-mandarin
- knoveleng/aishell1-mandarin
- knoveleng/aishell3-mandarin
- knoveleng/fleurs-mandarin
- knoveleng/fleurs-tamil
- knoveleng/slr127-tamil
- knoveleng/slr65-tamil
- knoveleng/cv-tamil
- knoveleng/mesolitica-malay
- knoveleng/fleurs-malay
- knoveleng/nsc-singlish
- knoveleng/librispeech-english
language:
- en
- zh
- ms
- ta
license: mit
pipeline_tag: automatic-speech-recognition
tags:
- singapore
- multilingual
- asr
- speech
- mlx
- speech-to-text
- speech
- transcription
- asr
- stt
- mlx-audio
library_name: mlx-audio
thucdangvan020999/polyglot-lion-1.7b-3bit
This model was converted to MLX format from knoveleng/polyglot-lion-1.7b using mlx-audio version 0.4.3.
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 thucdangvan020999/polyglot-lion-1.7b-3bit --audio "audio.wav"
Python Example:
from mlx_audio.stt.utils import load_model
from mlx_audio.stt.generate import generate_transcription
model = load_model("thucdangvan020999/polyglot-lion-1.7b-3bit")
transcription = generate_transcription(
model=model,
audio_path="path_to_audio.wav",
output_path="path_to_output.txt",
format="txt",
verbose=True,
)
print(transcription.text)