A fine-tuned wav2vec2.0/Hubert benchmark for SER, Speaker verification and spoken language understanding

Journal/Conference: arXiv preprint arXiv:2111.02735Year(published year): 2022Author: Yingzhi Wang, Abdelmoumene Boumadane, Abdelwahab HebaSubject: wav2vec 2.0, HuBERT, speech emotion recognition,