|
Hokkaido University Collection of Scholarly and Academic Papers >
Hokkaido University Sustainability Weeks >
Sustainability Weeks 2009 >
2009 APSIPA Annual Summit and Conference >
Regression Optimized Kernel for High-Level Speaker Verification
| Title: | Regression Optimized Kernel for High-Level Speaker Verification |
| Authors: | Zhang, Shi-Xiong Browse this author | | Mak, Man-Wai Browse this author |
| Keywords: | Speaker verification | | optimal kernels | | articulatory features | | pronunciation models | | SVM |
| Issue Date: | 4-Oct-2009 |
| Publisher: | Asia-Pacific Signal and Information Processing Association, 2009 Annual Summit and Conference, International Organizing Committee |
| Journal Title: | Proceedings : APSIPA ASC 2009 : Asia-Pacific Signal and Information Processing Association, 2009 Annual Summit and Conference |
| Start Page: | 40 |
| End Page: | 44 |
| Abstract: | Computing the likelihood-ratio (LR) score of a test utterance is an important step in speaker verification. It has recently been shown that for discrete speaker models, the LR scores can be expressed as dot products between supervectors formed by the test utterance, target-speaker model, and background model. This paper leverages this dot-product formulation and the representer theorem to derive a general kernel, namely the regression optimized kernel, for computing utterance-based verification scores using support vector machines. The kernel is general in that it can be a linear combination of any kernels belonging to the reproduction kernel Hilbert space. The combination weights are obtained by maximizing the ability of a discriminant function in separating a target speaker from impostors. The regression optimized kernel was applied to high-level speaker verification using articulatory-feature based pronunciation models. Results show that the scores produced by the regression optimized kernel are not only superior but also complementary to the LR scores, resulting in better performance when the two types of scores are combined. The proposed regression optimized kernel can be easily applied to other SVM-based classification problems. |
| Description: | APSIPA ASC 2009: Asia-Pacific Signal and Information Processing Association, 2009 Annual Summit and Conference. 4-7 October 2009. Sapporo, Japan. Oral session: Speech and Music Processing (5 October 2009). |
| Conference Name: | APSIPA ASC 2009: Asia-Pacific Signal and Information Processing Association, 2009 Annual Summit and Conference | | 2009年アジア太平洋信号情報処理連合学会アニュアルサミット・国際会議 |
| Conference Place: | Sapporo |
| Type: | proceedings |
| URI: | http://hdl.handle.net/2115/39640 |
| Appears in Collections: | 北海道大学サステナビリティ・ウィーク2009 (Sustainability Weeks 2009) > 2009年アジア太平洋信号情報処理連合学会アニュアルサミット・国際会議 (2009 APSIPA Annual Summit and Conference)
|
|
