Loading...
Please wait, while we are loading the content...
Similar Documents
Kernel Eigenspace-Based Mllr Adaptation Using Multiple Regression Classes (2005)
| Content Provider | CiteSeerX |
|---|---|
| Author | Hsiao, Roger Mak, Brian |
| Description | In Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing Recently, we have been investigating the application of kernel methods to improve the performance of eigenvoice-based adaptation methods by exploiting possible nonlinearity in their original working space. We proposed the kernel eigenvoice adaptation (KEV) in [1], and the kernel eigenspace-based MLLR adaptation (KEMLLR) in [2]. In KEMLLR, speaker-dependent MLLR transformation matrices are mapped to a kernel-induced high dimensional feature space, and kernel principal component analysis (KPCA) is used to derive a set of eigenmatrices in the feature space. A new speaker is then represented by a linear combination of the leading eigenmatrices. In this paper, we further improve KEMLLR by the use of multiple regression classes and the quasiNewton BFGS optimization algorithm. |
| File Format | |
| Language | English |
| Publisher Date | 2005-01-01 |
| Access Restriction | Open |
| Subject Keyword | Feature Space New Speaker Speaker-dependent Mllr Transformation Matrix Kernel Eigenspace-based Mllr Adaptation Quasinewton Bfgs Optimization Algorithm Possible Nonlinearity Leading Eigenmatrices Multiple Regression Class Kernel Method Eigenvoice-based Adaptation Method Original Working Space Kernel Eigenvoice Adaptation Kernel-induced High Dimensional Feature Space Principal Component Analysis Linear Combination |
| Content Type | Text |
| Resource Type | Article |
