Modeling uncertain speech sequences using type-2 fuzzy hidden markov models

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)peer-review

2 Scopus Citations
View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Title of host publicationAdvances in Multimedia Information Processing - PCM 2007
Subtitle of host publication8th Pacific Rim Conference on Multimedia, Proceedings
PublisherSpringer Verlag
Pages315-324
Volume4810 LNCS
ISBN (Print)9783540772545
Publication statusPublished - 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4810 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Title8th Pacific-Rim Conference on Multimedia (PCM 2007)
LocationCity University of Hong Kong
PlaceHong Kong
Period11 - 14 December 2007

Abstract

The automatic speech recognizor (ASR) based on hidden Markov models (HMMs) is very sensitive to multi-talker, non-stationary babble noise, which consists of a large number of speakers talking simultaneously. One major reason is due to mismatches between the training and testing conditions, which makes the accurate parameters of the HMM incapable of describing the uncertain distributions of the observations in speech signals. This paper applies one extension of the HMM referred to as the type-2 fuzzy hidden Markov models (T2 FHMMs) to modeling uncertain speech sequences. More specifically, we use the type2 fuzzy set (T2 FS) to describe uncertain parameters of the HMM that may vary anywhere in an interval with uniform possibilities. As a result, the likelihood of the T2 FHMM becomes an interval rather than a precise real number, which can be processed by the generalized linear model (GLM) for final classification decision-making. Experimental results of phoneme classification in the babble noise demonstrate a significant improvement compared with the HMM in terms of the robustness and classification rate. © Springer-Verlag Berlin Heidelberg 2007.

Research Area(s)

  • Babble noise, Hidden markov models, Type-2 fuzzy sets, Uncertain speech sequences

Citation Format(s)

Modeling uncertain speech sequences using type-2 fuzzy hidden markov models. / Cao, Xiao-Qin; Zeng, Jia; Yan, Hong.

Advances in Multimedia Information Processing - PCM 2007: 8th Pacific Rim Conference on Multimedia, Proceedings. Vol. 4810 LNCS Springer Verlag, 2007. p. 315-324 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4810 LNCS).

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)peer-review