Views
No views yet
python -m pip install faster-whisper==1.2.11from faster_whisper import WhisperModel
2
3model = WhisperModel(
4 "YOUR_NAMESPACE/Breeze-ASR-25-int8-CT2",
5 device="cpu",
6 compute_type="int8",
7)
8
9segments, info = model.transcribe(
10 "audio.wav",
11 language="zh",
12 beam_size=5,
13)
14
15print(f"language={info.language} probability={info.language_probability:.3f}")
16for segment in segments:
17 print(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}")device 改成 cuda；支援 FP16 的 GPU 可搭配 compute_type="int8_float16"。transformers.AutoModel* 直接載入。1@article{chou2025selfrefiningframeworkenhancingasr,
2 title={A Self-Refining Framework for Enhancing ASR Using TTS-Synthesized Data},
3 author={Cheng Kang Chou and Chan-Jan Hsu and Ho-Lam Chung and Liang-Hsuan Tseng and Hsi-Chun Cheng and Yu-Kuan Fu and Kuan Po Huang and Hung-Yi Lee},
4 journal={arXiv preprint arXiv:2506.11130},
5 year={2025}
6}