Continuous digit recognition in noise: Reservoirs can do an excellent job!

Azarakhsh Jalalvand; Fabian Triefenbach; Jean Pierre Martens

Continuous digit recognition in noise: Reservoirs can do an excellent job!

Azarakhsh Jalalvand
, Fabian Triefenbach
, Jean Pierre Martens

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper a formerly proposed continuous digit recognition system based on Reservoir Computing (RC) is improved in two respects: (1) the single reservoir is substituted by a stack of reservoirs, and (2) the straightforward mapping of reservoir outputs to state likelihoods is replaced by a trained non-parametric mapping. Furthermore, it is shown that a reservoir-based method can improve a model trained on clean speech to work better in a noisy condition from which it has a number of unknown digit string recordings available. The first two improvements have lead to a system that outperforms a HMM based system with the same noise robust features as input. The model adaptation offers a promising supplementary gain at modest noise levels.

Original language	English (US)
Title of host publication	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Pages	1802-1805
Number of pages	4
State	Published - 2012
Externally published	Yes
Event	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 - Portland, OR, United States Duration: Sep 9 2012 → Sep 13 2012

Publication series

Name	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Volume	3

Conference

Conference	13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012
Country/Territory	United States
City	Portland, OR
Period	9/9/12 → 9/13/12

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Communication

Keywords

Acoustic modeling
Model adaptation
Noise robustness
Reservoir computing

Cite this

@inproceedings{05df1b6faa6b414690b9e4a8971cc7fc,

title = "Continuous digit recognition in noise: Reservoirs can do an excellent job!",

abstract = "In this paper a formerly proposed continuous digit recognition system based on Reservoir Computing (RC) is improved in two respects: (1) the single reservoir is substituted by a stack of reservoirs, and (2) the straightforward mapping of reservoir outputs to state likelihoods is replaced by a trained non-parametric mapping. Furthermore, it is shown that a reservoir-based method can improve a model trained on clean speech to work better in a noisy condition from which it has a number of unknown digit string recordings available. The first two improvements have lead to a system that outperforms a HMM based system with the same noise robust features as input. The model adaptation offers a promising supplementary gain at modest noise levels.",

keywords = "Acoustic modeling, Model adaptation, Noise robustness, Reservoir computing",

author = "Azarakhsh Jalalvand and Fabian Triefenbach and Martens, \{Jean Pierre\}",

year = "2012",

language = "English (US)",

isbn = "9781622767595",

series = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

pages = "1802--1805",

booktitle = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012",

note = "13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012 ; Conference date: 09-09-2012 Through 13-09-2012",

}

Jalalvand, A, Triefenbach, F & Martens, JP 2012, Continuous digit recognition in noise: Reservoirs can do an excellent job! in 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, vol. 3, pp. 1802-1805, 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Portland, OR, United States, 9/9/12.

Continuous digit recognition in noise: Reservoirs can do an excellent job! / Jalalvand, Azarakhsh; Triefenbach, Fabian; Martens, Jean Pierre.
13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012. 2012. p. 1802-1805 (13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012; Vol. 3).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Continuous digit recognition in noise

T2 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

AU - Jalalvand, Azarakhsh

AU - Triefenbach, Fabian

AU - Martens, Jean Pierre

PY - 2012

Y1 - 2012

N2 - In this paper a formerly proposed continuous digit recognition system based on Reservoir Computing (RC) is improved in two respects: (1) the single reservoir is substituted by a stack of reservoirs, and (2) the straightforward mapping of reservoir outputs to state likelihoods is replaced by a trained non-parametric mapping. Furthermore, it is shown that a reservoir-based method can improve a model trained on clean speech to work better in a noisy condition from which it has a number of unknown digit string recordings available. The first two improvements have lead to a system that outperforms a HMM based system with the same noise robust features as input. The model adaptation offers a promising supplementary gain at modest noise levels.

AB - In this paper a formerly proposed continuous digit recognition system based on Reservoir Computing (RC) is improved in two respects: (1) the single reservoir is substituted by a stack of reservoirs, and (2) the straightforward mapping of reservoir outputs to state likelihoods is replaced by a trained non-parametric mapping. Furthermore, it is shown that a reservoir-based method can improve a model trained on clean speech to work better in a noisy condition from which it has a number of unknown digit string recordings available. The first two improvements have lead to a system that outperforms a HMM based system with the same noise robust features as input. The model adaptation offers a promising supplementary gain at modest noise levels.

KW - Acoustic modeling

KW - Model adaptation

KW - Noise robustness

KW - Reservoir computing

UR - https://www.scopus.com/pages/publications/84878591993

UR - https://www.scopus.com/pages/publications/84878591993#tab=citedBy

M3 - Conference contribution

AN - SCOPUS:84878591993

SN - 9781622767595

T3 - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

SP - 1802

EP - 1805

BT - 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012

Y2 - 9 September 2012 through 13 September 2012

ER -

Continuous digit recognition in noise: Reservoirs can do an excellent job!

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Keywords

Other files and links

Fingerprint

Cite this