Browse > Article

Improvements in Speaker Adaptation Using Weighted Training  

장규철 (한국과학기술원 전자전산학과)
우수영 (한국과학기술원 전자전산학과)
진민호 (한국과학기술원 전자전산학과)
박용규 (한국과학기술원 전자전산학과)
유창동 (한국과학기술원 전자전산학과)
Abstract
Regardless of the distribution of the adaptation data in the testing environment, model-based adaptation methods that have so far been reported in various literature incorporates the adaptation data undiscriminatingly in reducing the mismatch between the training and testing environments. When the amount of data is small and the parameter tying is extensive, adaptation based on outlier data can be detrimental to the performance of the recognizer. The distribution of the adaptation data plays a critical role on the adaptation performance. In order to maximally improve the recognition rate in the testing environment using only a small number of adaptation data, supervised weighted training is applied to the structural maximum a posterior (SMAP) algorithm. We evaluate the performance of the proposed weighted SMAP (WSMAP) and SMAP on TIDIGITS corpus. The proposed WSMAP has been found to perform better for a small amount of data. The general idea of incorporating the distribution of the adaptation data is applicable to other adaptation algorithms.
Keywords
Speaker adaptation; SMAP; WSMAP;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Structural MAP speaker adaptation using hierachical priors /
[ K.Shinoda;C.H.Lee ] / Proc. IEEE Workshop Speech Recognition Understanding
2 Efficient joint compensation of speech for the effects of additive noise and linear filtering /
[ F.H.Liu;A.Acero;R.Stern ] / IEEE International Conference on Acoustics, Speech, and Signal Processing, 1-257- 1-260
3 Hidden Markov model adaptation using maximum a posteriori linear regression /
[ O.Siohan;C.Chesta;C.H.Lee ] / Proc. Workshop Robust Methods for Speech Recognition in Adverse Conditions
4 An inequality with applications to statistical estimation for probabilistic functions of Markov processes and to a model for ecology /
[ L.E.Baum;J.A.Eagon ] / Bull. Amer. Math. Soc
5 On stochastic feature and model compensation approaches to robust speech recognition /
[ C.H.Lee ] / Speech Commun.   ScienceOn
6 Unsupervised speaker adaptation method based on hierarchical spectral clustering /
[ S.Furui ] / IEEE Trans. Acoust. Speech, Signal Processing   ScienceOn
7 A database for speaker-independent digit recognition /
[ R.G.Leonard ] / ICASSP
8 Discriminative learning for minimum error classification /
[ B.H.Juang;S.Katagirl ] / IEEE Trans. Signal Processing   ScienceOn
9 Selective training for hidden markov models with applications to speech classification /
[ L.M.Arsian;J.H.L.Hansen ] / IEEE Trans. Speech and Audio Processing   ScienceOn
10 Maximum a posteriori estimation for multivariate Gaussian mixture observations of markov chains /
[ Q.Huo;C.H.Lee ] / IEEE Trans. Speech Audio Processing   ScienceOn
11 Vector-field smoothed Bayesian learning for incremental speaker adaptation /
[ J.I.Takahashi;S.Sagayama ] / Proc. ICASSP-95
12 Maximum likelihood linear regression for speaker adaptation of continuous-density hidden markov models /
[ C.J.Leggetter;P.C.Woodland ] / Comput. Speech Lang.   ScienceOn