Brian Kan-Wing Mak

Publication List Details

Period

1989 - 2007

Number

17

Co-Authors

Kernel eigenspace-based MLLR adaptation (2007)

Mak, Brian Kan-Wing, Hsiao, Roger

Recently, we have been investigating the application of kernel methods for fast speaker adaptation by exploiting possible non-linearity in the input speaker space. In this paper, we propose another...

Robustness of several kernel-based fast adaptation methods on noisy LVCSR (2007)

Mak, Brian Kan-Wing, Hsiao, Roger

We have been investigating the use of kernel methods to improve conventional linear adaptation algorithms for fast adaptation, when there are less than 10s of adaptation speech. On clean speech, we...

Minimization of utterance verification error rate as a constrained optimization problem (2006)

Siu, Man Hung, Mak, Brian Kan-Wing, Au, Wing-hei

Since utterance verification (UV) may be treated as a 2-class classification problem, it may be improved with discriminative training such as minimum verification error training or minimum...

Joint optimization of the frequency-domain and time-domain transformations in deriving generalized static and dynamic MFCCs (2006)

Lai, Yiu-Pong, Siu, Man Hung, Mak, Brian Kan-Wing

Traditionally, static mel-frequency cepstral coefficients (MFCCs) are derived by discrete cosine transformation (DCT), and dynamic MFCCs are derived by linear regression. Their derivation may be...

Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting (2006)

Mak, Brian Kan-Wing, Hsiao, Roger, Ho, Simon, Kwok, James Tin-Yau

Recently, we proposed an improvement to theconventional eigenvoice (EV) speaker adaptation using kernel methods. In our novel kernel eigenvoice (KEV) speaker adaptation [1], speaker supervectors are...

Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting (2006)

Mak, Brian Kan-Wing, Hsiao, Roger, Ho, Simon, Kwok, James Tin-Yau

Recently, we proposed an improvement to theconventional eigenvoice (EV) speaker adaptation using kernel methods. In our novel kernel eigenvoice (KEV) speaker adaptation [1], speaker supervectors are...

High-density discrete HMM with the use of scalar quantization indexing (2005)

Mak, Brian Kan-Wing, Au Yeung, Siu-Kei, Lai, Yiu-Pong, Siu, Man Hung

With the advance in semiconductor memory and the availability of very large speech corpora (of hundreds to thousands of hours of speech), we would like to revisit the use of discrete hidden Markov...

A comparative study of two kernel eigenspace-based speaker adaptation methods on large vocabulary continuous speech recognition (2005)

Hsiao, Roger, Mak, Brian Kan-Wing

Eigenvoice (EV) speaker adaptation has been shown effective for fast speaker adaptation when the amount of adaptation data is scarce. In the past two years, we have been investigating the application...

Pruning hidden Markov models with optimal brain surgeon (2005)

Mak, Brian Kan-Wing, Chan, Kin-Wah

A method of pruning hidden Markov models (HMMs) is presented. The main purpose is to find a good HMM topology for a given task with improved generalization capability. As a side effect, the resulting...

Various reference speakers determination methods for embedded kernel eigenvoice speaker adaptation (2005)

Mak, Brian Kan-Wing, Ho, Simon

Recently, we proposed two improvements to the eigenvoice (EV) speaker adaptation using kernel methods: kernel eigenvoice (KEV) speaker adaptation, and embedded kernel eigenvoice (eKEV) speaker...

Kernel eigenspace-based MLLR adaptation using multiple regression classes (2005)

Hsiao, Roger, Mak, Brian Kan-Wing

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...

Speedup of kernel eigenvoice speaker adaptation by embedded kernel PCA (2004)

Mak, Brian Kan-Wing, Ho, Simon, Kwok, Tin-Yau

Recently, we proposed an improvement to the eigenvoice (EV) speaker adaptation called kernel eigenvoice (KEV) speaker adaptation. In KEV adaptation, eigenvoices are computed using kernel PCA, and a...

Improving eigenspace-based MLLR adaptation by kernel PCA (2004)

Mak, Brian Kan-Wing, Hsiao, Roger

Eigenspace-based MLLR (EMLLR) adaptation has been shown effective for fast speaker adaptation. It applies the basic idea of eigenvoice adaptation, and derives a small set of eigenmatrices using...