模型:

speechbrain/asr-wav2vec2-commonvoice-en