Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment

Soonho Baek, Hong Goo Kang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This paper proposes a novel HMM adaptation algorithm for robust automatic speech recognition (ASR) system in noisy environments. The HMM adaptation using vector Taylor series (VTS) significantly improves the ASR performance in noisy environments. Recently, the power normalized cepstral coefficient (PNCC) that replaces a logarithmic mapping function with a power mapping function has been proposed and it is proved that the replacement of the mapping function is robust to additive noise. In this paper, we extend the VTS based approach to the cepstral coefficients obtained by using a power mapping function instead of a logarithmic mapping function. Experimental results indicate that HMM adaptation in the cepstrum obtained by using a power mapping function improves the ASR performance comparing the VTS based conventional approach for mel-frequency cepstral coefficients (MFCCs).

Original languageEnglish
Title of host publication2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings
Pages186-191
Number of pages6
DOIs
Publication statusPublished - 2013 Dec 1
Event2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Olomouc, Czech Republic
Duration: 2013 Dec 82013 Dec 13

Publication series

Name2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings

Other

Other2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013
CountryCzech Republic
CityOlomouc
Period13/12/813/12/13

Fingerprint

Noise
Power (Psychology)

All Science Journal Classification (ASJC) codes

  • Speech and Hearing

Cite this

Baek, S., & Kang, H. G. (2013). Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment. In 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings (pp. 186-191). [6707727] (2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings). https://doi.org/10.1109/ASRU.2013.6707727
Baek, Soonho ; Kang, Hong Goo. / Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment. 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings. 2013. pp. 186-191 (2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings).
@inproceedings{d81902a0ee7a46dcb64621291a8f72ee,
title = "Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment",
abstract = "This paper proposes a novel HMM adaptation algorithm for robust automatic speech recognition (ASR) system in noisy environments. The HMM adaptation using vector Taylor series (VTS) significantly improves the ASR performance in noisy environments. Recently, the power normalized cepstral coefficient (PNCC) that replaces a logarithmic mapping function with a power mapping function has been proposed and it is proved that the replacement of the mapping function is robust to additive noise. In this paper, we extend the VTS based approach to the cepstral coefficients obtained by using a power mapping function instead of a logarithmic mapping function. Experimental results indicate that HMM adaptation in the cepstrum obtained by using a power mapping function improves the ASR performance comparing the VTS based conventional approach for mel-frequency cepstral coefficients (MFCCs).",
author = "Soonho Baek and Kang, {Hong Goo}",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ASRU.2013.6707727",
language = "English",
isbn = "9781479927562",
series = "2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings",
pages = "186--191",
booktitle = "2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings",

}

Baek, S & Kang, HG 2013, Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment. in 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings., 6707727, 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings, pp. 186-191, 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013, Olomouc, Czech Republic, 13/12/8. https://doi.org/10.1109/ASRU.2013.6707727

Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment. / Baek, Soonho; Kang, Hong Goo.

2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings. 2013. p. 186-191 6707727 (2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment

AU - Baek, Soonho

AU - Kang, Hong Goo

PY - 2013/12/1

Y1 - 2013/12/1

N2 - This paper proposes a novel HMM adaptation algorithm for robust automatic speech recognition (ASR) system in noisy environments. The HMM adaptation using vector Taylor series (VTS) significantly improves the ASR performance in noisy environments. Recently, the power normalized cepstral coefficient (PNCC) that replaces a logarithmic mapping function with a power mapping function has been proposed and it is proved that the replacement of the mapping function is robust to additive noise. In this paper, we extend the VTS based approach to the cepstral coefficients obtained by using a power mapping function instead of a logarithmic mapping function. Experimental results indicate that HMM adaptation in the cepstrum obtained by using a power mapping function improves the ASR performance comparing the VTS based conventional approach for mel-frequency cepstral coefficients (MFCCs).

AB - This paper proposes a novel HMM adaptation algorithm for robust automatic speech recognition (ASR) system in noisy environments. The HMM adaptation using vector Taylor series (VTS) significantly improves the ASR performance in noisy environments. Recently, the power normalized cepstral coefficient (PNCC) that replaces a logarithmic mapping function with a power mapping function has been proposed and it is proved that the replacement of the mapping function is robust to additive noise. In this paper, we extend the VTS based approach to the cepstral coefficients obtained by using a power mapping function instead of a logarithmic mapping function. Experimental results indicate that HMM adaptation in the cepstrum obtained by using a power mapping function improves the ASR performance comparing the VTS based conventional approach for mel-frequency cepstral coefficients (MFCCs).

UR - http://www.scopus.com/inward/record.url?scp=84893654412&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893654412&partnerID=8YFLogxK

U2 - 10.1109/ASRU.2013.6707727

DO - 10.1109/ASRU.2013.6707727

M3 - Conference contribution

AN - SCOPUS:84893654412

SN - 9781479927562

T3 - 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings

SP - 186

EP - 191

BT - 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings

ER -

Baek S, Kang HG. Vector Taylor series based HMM adaptation for generalized cepstrum in noisy environment. In 2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings. 2013. p. 186-191. 6707727. (2013 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2013 - Proceedings). https://doi.org/10.1109/ASRU.2013.6707727