TY - GEN
T1 - Adversarial Training for Multi-domain Speaker Recognition
AU - Wang, Qing
AU - Rao, Wei
AU - Guo, Pengcheng
AU - Xie, Lei
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/1/24
Y1 - 2021/1/24
N2 - In real-life applications, the performance of speaker recognition systems always degrades when there is a mismatch between training and evaluation data. Many domain adaptation methods have been successfully used for eliminating the domain mismatches in speaker recognition. However, usually both training and evaluation data themselves can be composed of several subsets. These inner variances of each dataset can also be considered as different domains. Different distributed subsets in source or target domain dataset can also cause multi-domain mismatches, which are influential to speaker recognition performance. In this study, we propose to use adversarial training for multi-domain speaker recognition to solve the domain mismatch and the dataset variance problems. By adopting the proposed method, we are able to obtain both multi-domain-invariant and speaker-discriminative speech representations for speaker recognition. Experimental results on DAC13 dataset indicate that the proposed method is not only effective to solve the multi-domain mismatch problem, but also outperforms the compared unsupervised domain adaptation methods.
AB - In real-life applications, the performance of speaker recognition systems always degrades when there is a mismatch between training and evaluation data. Many domain adaptation methods have been successfully used for eliminating the domain mismatches in speaker recognition. However, usually both training and evaluation data themselves can be composed of several subsets. These inner variances of each dataset can also be considered as different domains. Different distributed subsets in source or target domain dataset can also cause multi-domain mismatches, which are influential to speaker recognition performance. In this study, we propose to use adversarial training for multi-domain speaker recognition to solve the domain mismatch and the dataset variance problems. By adopting the proposed method, we are able to obtain both multi-domain-invariant and speaker-discriminative speech representations for speaker recognition. Experimental results on DAC13 dataset indicate that the proposed method is not only effective to solve the multi-domain mismatch problem, but also outperforms the compared unsupervised domain adaptation methods.
KW - adversarial training
KW - multi-domain adaptation
KW - speaker recognition
UR - http://www.scopus.com/inward/record.url?scp=85102572596&partnerID=8YFLogxK
U2 - 10.1109/ISCSLP49672.2021.9362053
DO - 10.1109/ISCSLP49672.2021.9362053
M3 - 会议稿件
AN - SCOPUS:85102572596
T3 - 2021 12th International Symposium on Chinese Spoken Language Processing, ISCSLP 2021
BT - 2021 12th International Symposium on Chinese Spoken Language Processing, ISCSLP 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th International Symposium on Chinese Spoken Language Processing, ISCSLP 2021
Y2 - 24 January 2021 through 27 January 2021
ER -