On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system

Shinji Watanabe, Atsushi Nakamura

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

3 Citations (Scopus)

Abstract

Acoustic characteristics are often changed over time as a result of various factors including changes of speakers, speaking styles, and noise sources. Incremental adaptation techniques for speech recognition are aimed at adjusting acoustic models quickly and stably to such time-variant acoustic characteristics. Recently we proposed a novel incremental adaptation framework based on a macroscopic time evolution system, which models the time-variant characteristics by successively updating posterior distributions of acoustic model parameters. This paper proposes fast incremental adaptation based on a macroscopic time evolution system that realizes an utterance-by-utterance update by approximating the posterior distributions. This adaptation was used to perform on-line adaptation of Japanese broadcast news for very large vocabulary continuous speech recognition (700k vocabulary size) in real time. The word accuracy was improved from 73.9% to 85.1%. In addition, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.

Original languageEnglish
Title of host publication2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009
Pages4373-4376
Number of pages4
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009 - Taipei
Duration: 2009 Apr 192009 Apr 24

Other

Other2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009
CityTaipei
Period09/4/1909/4/24

Fingerprint

Acoustics
Continuous speech recognition
Speech recognition
Acoustic noise

Keywords

  • Acoustic model
  • Macroscopic time evolution system
  • Model selection
  • On-line adaptation
  • Speech recognition

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Watanabe, S., & Nakamura, A. (2009). On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009 (pp. 4373-4376). [4960598] https://doi.org/10.1109/ICASSP.2009.4960598

On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system. / Watanabe, Shinji; Nakamura, Atsushi.

2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 4373-4376 4960598.

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

Watanabe, S & Nakamura, A 2009, On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system. in 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009., 4960598, pp. 4373-4376, 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009, Taipei, 09/4/19. https://doi.org/10.1109/ICASSP.2009.4960598
Watanabe S, Nakamura A. On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 4373-4376. 4960598 https://doi.org/10.1109/ICASSP.2009.4960598
Watanabe, Shinji ; Nakamura, Atsushi. / On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system. 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. pp. 4373-4376
@inproceedings{dd0f4239cb2641769da0b54b3f4099fb,
title = "On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system",
abstract = "Acoustic characteristics are often changed over time as a result of various factors including changes of speakers, speaking styles, and noise sources. Incremental adaptation techniques for speech recognition are aimed at adjusting acoustic models quickly and stably to such time-variant acoustic characteristics. Recently we proposed a novel incremental adaptation framework based on a macroscopic time evolution system, which models the time-variant characteristics by successively updating posterior distributions of acoustic model parameters. This paper proposes fast incremental adaptation based on a macroscopic time evolution system that realizes an utterance-by-utterance update by approximating the posterior distributions. This adaptation was used to perform on-line adaptation of Japanese broadcast news for very large vocabulary continuous speech recognition (700k vocabulary size) in real time. The word accuracy was improved from 73.9{\%} to 85.1{\%}. In addition, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.",
keywords = "Acoustic model, Macroscopic time evolution system, Model selection, On-line adaptation, Speech recognition",
author = "Shinji Watanabe and Atsushi Nakamura",
year = "2009",
doi = "10.1109/ICASSP.2009.4960598",
language = "English",
isbn = "9781424423545",
pages = "4373--4376",
booktitle = "2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009",

}

TY - GEN

T1 - On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system

AU - Watanabe, Shinji

AU - Nakamura, Atsushi

PY - 2009

Y1 - 2009

N2 - Acoustic characteristics are often changed over time as a result of various factors including changes of speakers, speaking styles, and noise sources. Incremental adaptation techniques for speech recognition are aimed at adjusting acoustic models quickly and stably to such time-variant acoustic characteristics. Recently we proposed a novel incremental adaptation framework based on a macroscopic time evolution system, which models the time-variant characteristics by successively updating posterior distributions of acoustic model parameters. This paper proposes fast incremental adaptation based on a macroscopic time evolution system that realizes an utterance-by-utterance update by approximating the posterior distributions. This adaptation was used to perform on-line adaptation of Japanese broadcast news for very large vocabulary continuous speech recognition (700k vocabulary size) in real time. The word accuracy was improved from 73.9% to 85.1%. In addition, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.

AB - Acoustic characteristics are often changed over time as a result of various factors including changes of speakers, speaking styles, and noise sources. Incremental adaptation techniques for speech recognition are aimed at adjusting acoustic models quickly and stably to such time-variant acoustic characteristics. Recently we proposed a novel incremental adaptation framework based on a macroscopic time evolution system, which models the time-variant characteristics by successively updating posterior distributions of acoustic model parameters. This paper proposes fast incremental adaptation based on a macroscopic time evolution system that realizes an utterance-by-utterance update by approximating the posterior distributions. This adaptation was used to perform on-line adaptation of Japanese broadcast news for very large vocabulary continuous speech recognition (700k vocabulary size) in real time. The word accuracy was improved from 73.9% to 85.1%. In addition, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.

KW - Acoustic model

KW - Macroscopic time evolution system

KW - Model selection

KW - On-line adaptation

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=70349213985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349213985&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2009.4960598

DO - 10.1109/ICASSP.2009.4960598

M3 - Conference contribution

AN - SCOPUS:70349213985

SN - 9781424423545

SP - 4373

EP - 4376

BT - 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009

ER -