模型:

speechbrain/emotion-recognition-wav2vec2-IEMOCAP