A Study on Invariance of $f$-Divergence and Its Application to Speech Recognition
Authors:
- 1 April 2010
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 58 (7) , 3884-3890
Abstract
Identifying features invariant to certain transformations is a fundamental problem in the fields of signal processing and pattern recognition. This correspondence explores a family of measures called f -divergences that are invariant to invertible transformations, and studies their application to speech recognition. We provide novel proofs for the sufficiency and necessity of the invariance of f -divergence. Several techniques to calculate or approximate f -divergences in general cases and for special distributions such as Gaussian and Gaussian mixture are reviewed. We show how to construct an invariant structural representation from sequence data through maximum likelihood decomposition, and prove the invariance of this decomposition. We demonstrate an application of this invariant representation to recognizing connected Japanese vowel utterances. In addition, we propose several techniques to improve the recognition performance. The experimental results show that the invariant structure achieves better performance than hidden Markov models, a widely used technique for acoustic modeling of speech sounds.Keywords
This publication has 18 references indexed in Scilit:
- Automatic speech recognition and speech variability: A reviewSpeech Communication, 2007
- Mathematical Evidence of the Acoustic Universal Structure in Speech2006
- On Divergences and Informations in Statistics and Information TheoryIEEE Transactions on Information Theory, 2006
- STRUCTURAL REPRESENTATION OF THE PRONUNCIATION AND ITS USE FOR CALL2006
- Vocal tract normalization equals linear transformation in cepstral spaceIEEE Transactions on Speech and Audio Processing, 2005
- Information Theory and Statistics: A TutorialFoundations and Trends® in Communications and Information Theory, 2004
- Yet another acoustic representation of speech sounds2004
- Fast approximation of Kullback-Leibler distance for dependence trees and hidden Markov modelsIEEE Signal Processing Letters, 2003
- About distances of discrete distributions satisfying the data processing theorem of information theoryIEEE Transactions on Information Theory, 1997
- On thef-divergence and singularity of probability measuresPeriodica Mathematica Hungarica, 1972