Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition

Roy D. Patterson; Thomas C. Walters; Jessica Monaghan; Christian Feldbauer; Toshio Irino

doi:10.1109/ISCAS.2010.5537725

Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition

Roy D. Patterson, Thomas C. Walters, Jessica Monaghan, Christian Feldbauer, Toshio Irino

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

1 Citation (Scopus)

Abstract

The syllables of speech contain information about the vocal tract length (VTL) of the speaker as well as the phonetic message. Ideally, the pre-processor used for automatic speech recognition (ASR) should segregate the phonetic message from the VTL information. This paper describes a method to calculate VTL-invariant auditory feature vectors from speech, using a method in which the message and the VTL are segregated. Spectra produced by an auditory filterbank are summarized by a Gaussian mixture model (GMM) to produce a low-dimensional feature vector. These features are evaluated for robustness in comparison with conventional mel-frequency cepstral coefficients (MFCCs) using a hidden-Markov-model (HMM) recognizer. A dynamic, compressive gammachirp (dcGC) auditory filterbank is also introduced. The dcGC provides a leveldependent spectral analysis, with near instantaneous compression, and two-tone suppression.

Original language	English
Title of host publication	ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems
Place of Publication	Piscataway, NJ
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	3813-3816
Number of pages	4
ISBN (Print)	9781424453085
DOIs	https://doi.org/10.1109/ISCAS.2010.5537725
Publication status	Published - 2010
Externally published	Yes
Event	2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, ISCAS 2010 - Paris, France Duration: 30 May 2010 → 2 Jun 2010

Other

Other	2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, ISCAS 2010
Country/Territory	France
City	Paris
Period	30/05/10 → 2/06/10

Access to Document

10.1109/ISCAS.2010.5537725

Cite this

Patterson, R. D., Walters, T. C., Monaghan, J., Feldbauer, C., & Irino, T. (2010). Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition. In ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems (pp. 3813-3816). Article 5537725 Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ISCAS.2010.5537725

Patterson, Roy D. ; Walters, Thomas C. ; Monaghan, Jessica et al. / Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition. ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems. Piscataway, NJ : Institute of Electrical and Electronics Engineers (IEEE), 2010. pp. 3813-3816

@inproceedings{9290dc0e199b4b11ac60ac51546734a2,

title = "Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition",

abstract = "The syllables of speech contain information about the vocal tract length (VTL) of the speaker as well as the phonetic message. Ideally, the pre-processor used for automatic speech recognition (ASR) should segregate the phonetic message from the VTL information. This paper describes a method to calculate VTL-invariant auditory feature vectors from speech, using a method in which the message and the VTL are segregated. Spectra produced by an auditory filterbank are summarized by a Gaussian mixture model (GMM) to produce a low-dimensional feature vector. These features are evaluated for robustness in comparison with conventional mel-frequency cepstral coefficients (MFCCs) using a hidden-Markov-model (HMM) recognizer. A dynamic, compressive gammachirp (dcGC) auditory filterbank is also introduced. The dcGC provides a leveldependent spectral analysis, with near instantaneous compression, and two-tone suppression.",

author = "Patterson, \{Roy D.\} and Walters, \{Thomas C.\} and Jessica Monaghan and Christian Feldbauer and Toshio Irino",

year = "2010",

doi = "10.1109/ISCAS.2010.5537725",

language = "English",

isbn = "9781424453085",

pages = "3813--3816",

booktitle = "ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, ISCAS 2010 ; Conference date: 30-05-2010 Through 02-06-2010",

}

Patterson, RD, Walters, TC, Monaghan, J, Feldbauer, C & Irino, T 2010, Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition. in ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems., 5537725, Institute of Electrical and Electronics Engineers (IEEE), Piscataway, NJ, pp. 3813-3816, 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, ISCAS 2010, Paris, France, 30/05/10. https://doi.org/10.1109/ISCAS.2010.5537725

Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition. / Patterson, Roy D.; Walters, Thomas C.; Monaghan, Jessica et al.
ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems. Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE), 2010. p. 3813-3816 5537725.

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

TY - GEN

T1 - Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition

AU - Patterson, Roy D.

AU - Walters, Thomas C.

AU - Monaghan, Jessica

AU - Feldbauer, Christian

AU - Irino, Toshio

PY - 2010

Y1 - 2010

N2 - The syllables of speech contain information about the vocal tract length (VTL) of the speaker as well as the phonetic message. Ideally, the pre-processor used for automatic speech recognition (ASR) should segregate the phonetic message from the VTL information. This paper describes a method to calculate VTL-invariant auditory feature vectors from speech, using a method in which the message and the VTL are segregated. Spectra produced by an auditory filterbank are summarized by a Gaussian mixture model (GMM) to produce a low-dimensional feature vector. These features are evaluated for robustness in comparison with conventional mel-frequency cepstral coefficients (MFCCs) using a hidden-Markov-model (HMM) recognizer. A dynamic, compressive gammachirp (dcGC) auditory filterbank is also introduced. The dcGC provides a leveldependent spectral analysis, with near instantaneous compression, and two-tone suppression.

AB - The syllables of speech contain information about the vocal tract length (VTL) of the speaker as well as the phonetic message. Ideally, the pre-processor used for automatic speech recognition (ASR) should segregate the phonetic message from the VTL information. This paper describes a method to calculate VTL-invariant auditory feature vectors from speech, using a method in which the message and the VTL are segregated. Spectra produced by an auditory filterbank are summarized by a Gaussian mixture model (GMM) to produce a low-dimensional feature vector. These features are evaluated for robustness in comparison with conventional mel-frequency cepstral coefficients (MFCCs) using a hidden-Markov-model (HMM) recognizer. A dynamic, compressive gammachirp (dcGC) auditory filterbank is also introduced. The dcGC provides a leveldependent spectral analysis, with near instantaneous compression, and two-tone suppression.

UR - https://www.scopus.com/pages/publications/77955991370

U2 - 10.1109/ISCAS.2010.5537725

DO - 10.1109/ISCAS.2010.5537725

M3 - Conference proceeding contribution

AN - SCOPUS:77955991370

SN - 9781424453085

SP - 3813

EP - 3816

BT - ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - Piscataway, NJ

T2 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, ISCAS 2010

Y2 - 30 May 2010 through 2 June 2010

ER -

Patterson RD, Walters TC, Monaghan J, Feldbauer C, Irino T. Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition. In ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems. Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE). 2010. p. 3813-3816. 5537725 doi: 10.1109/ISCAS.2010.5537725

Auditory speech processing for scale-shift covariance and its evaluation in automatic speech recognition

Abstract

Other

Access to Document

Other files and links

Fingerprint

Cite this