Speaker adaptation in speech recognition using linear regression techniques

Stephen J. Cox

doi:10.1049/el:19921342

Speaker adaptation in speech recognition using linear regression techniques

Stephen J. Cox

Research output: Contribution to journal › Article

3 Citations (Scopus)

Abstract

A new technique of speaker adaptation for use in speaker-independent speech recognition systems is presented. The training data is used to build models (based on linear regression) of sounds. At recognition time, the models are used together with an incomplete set of sounds from a new speaker to estimate values for unheard sounds, which are used to adapt the speaker-independent models.

Original language	English
Pages (from-to)	2093-2094
Number of pages	2
Journal	IEE Electronics Letters
Volume	28
Issue number	22
DOIs	https://doi.org/10.1049/el:19921342
Publication status	Published - Oct 1992

Access to Document

10.1049/el:19921342

Cite this

@article{5ec92ca2578d44b7a9d46fdeed98983d,

title = "Speaker adaptation in speech recognition using linear regression techniques",

abstract = "A new technique of speaker adaptation for use in speaker-independent speech recognition systems is presented. The training data is used to build models (based on linear regression) of sounds. At recognition time, the models are used together with an incomplete set of sounds from a new speaker to estimate values for unheard sounds, which are used to adapt the speaker-independent models.",

author = "Cox, {Stephen J.}",

year = "1992",

month = oct,

doi = "10.1049/el:19921342",

language = "English",

volume = "28",

pages = "2093--2094",

journal = "IEE Electronics Letters",

issn = "0013-5194",

publisher = "Institution of Engineering and Technology",

number = "22",

}

TY - JOUR

T1 - Speaker adaptation in speech recognition using linear regression techniques

AU - Cox, Stephen J.

PY - 1992/10

Y1 - 1992/10

N2 - A new technique of speaker adaptation for use in speaker-independent speech recognition systems is presented. The training data is used to build models (based on linear regression) of sounds. At recognition time, the models are used together with an incomplete set of sounds from a new speaker to estimate values for unheard sounds, which are used to adapt the speaker-independent models.

AB - A new technique of speaker adaptation for use in speaker-independent speech recognition systems is presented. The training data is used to build models (based on linear regression) of sounds. At recognition time, the models are used together with an incomplete set of sounds from a new speaker to estimate values for unheard sounds, which are used to adapt the speaker-independent models.

U2 - 10.1049/el:19921342

DO - 10.1049/el:19921342

M3 - Article

VL - 28

SP - 2093

EP - 2094

JO - IEE Electronics Letters

JF - IEE Electronics Letters

SN - 0013-5194

IS - 22

ER -