ISCA Archive Interspeech 2006
ISCA Archive Interspeech 2006

Study on speaker verification on emotional speech

Wei Wu, Thomas Fang Zheng, Ming-Xing Xu, Huan-Jun Bao

Besides background noise, channel effect and speaker’s health condition, emotion is another factor which may influence the performance of a speaker verification system. In this paper, the performance of a GMM-UBM based speaker verification system on emotional speech is studied. It is found that speech with various emotions aggravates the verification performance. Two reasons for the performance aggravation are analyzed, they are mismatched emotions between the speaker models and the test utterances, and the articulating styles of certain emotions which create intense intra-speaker vocal variability. In response to the first reason, an emotion-dependent score normalization method is proposed, which is borrowed from the idea of Hnorm.

doi: 10.21437/Interspeech.2006-191

Cite as: Wu, W., Zheng, T.F., Xu, M.-X., Bao, H.-J. (2006) Study on speaker verification on emotional speech. Proc. Interspeech 2006, paper 1124-Wed3CaP.7, doi: 10.21437/Interspeech.2006-191

  author={Wei Wu and Thomas Fang Zheng and Ming-Xing Xu and Huan-Jun Bao},
  title={{Study on speaker verification on emotional speech}},
  booktitle={Proc. Interspeech 2006},
  pages={paper 1124-Wed3CaP.7},