DESCRIPTION OF ACOUSTIC VARIATIONS BY TREE-BASED PHONE MODELING

Satoru Hayamizu, Kai Fu Lee, Hsiao Wuen Hon

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)

Abstract

This paper discusses the use of tree-based phone modeling to describe acoustic variations of speech, and its application to speech recognition system. There are many sources of variabilities that affect the realization of a phoneme: phonetic contexts, speakers, stress, speaking rates and so on. Explicit modeling with these sources of variabilities will give more accurate and more detailed phone models, but needs a large amount of speech data for training. Tree-based phone modeling is studied to solve this problem with three case studies: phone models with large VQ codebook sizes, decision tree clustering, and speaker-clustering. They are tested on speakerindependent continuous speech recognition experiments with a 991 word vocabulary. Tree-based phone modeling is shown to produce improvement in all three cases and to provide a good guide to provide trainability and generalizability.

Original languageEnglish
Pages705-708
Number of pages4
Publication statusPublished - 1990
Event1st International Conference on Spoken Language Processing, ICSLP 1990 - Kobe, Japan
Duration: 1990 Nov 181990 Nov 22

Conference

Conference1st International Conference on Spoken Language Processing, ICSLP 1990
Country/TerritoryJapan
CityKobe
Period90/11/1890/11/22

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'DESCRIPTION OF ACOUSTIC VARIATIONS BY TREE-BASED PHONE MODELING'. Together they form a unique fingerprint.

Cite this