Speaker verification based on the fusion of speech acoustics and inverted articulatory signals
详细信息 查看全文
- 作者：Ming Lia ; b ; c ; liming46@mail.sysu.edu.cn" class="auth_mail" title="E-mail the corresponding author ; ; Jangwon Kimd ; jangwon@usc.edu" class="auth_mail" title="E-mail the corresponding author ; ; Adam Lammertd ; lammert@usc.edu" class="auth_mail" title="E-mail the corresponding author ; ; Prasanta Kumar Ghoshe ; prasantg@ee.iisc.ernet.in" class="auth_mail" title="E-mail the corresponding author ; ; Vikram Ramanarayanand ; vramanar@usc.edu" class="auth_mail" title="E-mail the corresponding author ; ; Shrikanth Narayanand ; shri@sipi.usc.edu" class="auth_mail" title="E-mail the corresponding author ;
- 关键词：Text independent speaker verification ; Text dependent speaker verification ; Speech production ; Articulatory features ; Acoustic-to-articulatory inversion
- 刊名：Computer Speech & Language
- 出版年：2016
- 出版时间：March 2016
- 年：2016
- 卷：36
- 期：Complete
- 页码：196-211
- 全文大小：2465 K
文摘
- •
A practical feature-level and score-level fusion approach for speaker verification with articulatory information.
- •
Concatenating real articulatory measurements with MFCCs improves the performance.
- •
Concatenating acoustic-to-articulatory inversion features with MFCCs also improves the result.