Feature extraction with a multiscale modulation analysis for robust automatic speech recognition

F. Müller; A. Mertins

doi:10.1109/ICASSP.2013.6639106

Feature extraction with a multiscale modulation analysis for robust automatic speech recognition

Institute of Signal Processing

Abstract

In this work we present a new feature extraction method that is robust against the effects of varying vocal tract lengths. The principle of the method is based on invariant integration and makes use of a modulation filtering approach, similar to the recently proposed scattering transform. In particular, we show how the transform can be used to obtain features that are robust against variations of the vocal tract length. Phoneme recognition experiments show a clearly increased robustness in case of mismatching average vocal tract lengths.

Original language	English
Title of host publication	2013 IEEE International Conference on Acoustics, Speech and Signal Processing
Number of pages	5
Publisher	IEEE
Publication date	01.05.2013
Pages	7427-7431
Article number	6639106
ISBN (Electronic)	978-1-4799-0356-6
DOIs	https://doi.org/10.1109/ICASSP.2013.6639106
Publication status	Published - 01.05.2013
Event	2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing - Vancouver, Canada Duration: 26.05.2013 → 31.05.2013 Conference number: 101421

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICASSP.2013.6639106

Cite this

Müller, F., & Mertins, A. (2013). Feature extraction with a multiscale modulation analysis for robust automatic speech recognition. In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 7427-7431). Article 6639106 (IEEE International Conference on Acoustics, Speech and Signal Processing). IEEE. https://doi.org/10.1109/ICASSP.2013.6639106

@inproceedings{bc982c7be385440caf296060a3965a10,

title = "Feature extraction with a multiscale modulation analysis for robust automatic speech recognition",

abstract = "In this work we present a new feature extraction method that is robust against the effects of varying vocal tract lengths. The principle of the method is based on invariant integration and makes use of a modulation filtering approach, similar to the recently proposed scattering transform. In particular, we show how the transform can be used to obtain features that are robust against variations of the vocal tract length. Phoneme recognition experiments show a clearly increased robustness in case of mismatching average vocal tract lengths.",

author = "F. M{\"u}ller and A. Mertins",

year = "2013",

month = may,

day = "1",

doi = "10.1109/ICASSP.2013.6639106",

language = "English",

series = "IEEE International Conference on Acoustics, Speech and Signal Processing",

publisher = "IEEE",

pages = "7427--7431",

booktitle = "2013 IEEE International Conference on Acoustics, Speech and Signal Processing",

note = "2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 ; Conference date: 26-05-2013 Through 31-05-2013",

}

Müller, F & Mertins, A 2013, Feature extraction with a multiscale modulation analysis for robust automatic speech recognition. in 2013 IEEE International Conference on Acoustics, Speech and Signal Processing., 6639106, IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE, pp. 7427-7431, 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, Vancouver, British Columbia, Canada, 26.05.13. https://doi.org/10.1109/ICASSP.2013.6639106

Feature extraction with a multiscale modulation analysis for robust automatic speech recognition. / Müller, F.; Mertins, A.
2013 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 2013. p. 7427-7431 6639106 (IEEE International Conference on Acoustics, Speech and Signal Processing).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - Feature extraction with a multiscale modulation analysis for robust automatic speech recognition

AU - Müller, F.

AU - Mertins, A.

N1 - Conference code: 101421

PY - 2013/5/1

Y1 - 2013/5/1

N2 - In this work we present a new feature extraction method that is robust against the effects of varying vocal tract lengths. The principle of the method is based on invariant integration and makes use of a modulation filtering approach, similar to the recently proposed scattering transform. In particular, we show how the transform can be used to obtain features that are robust against variations of the vocal tract length. Phoneme recognition experiments show a clearly increased robustness in case of mismatching average vocal tract lengths.

AB - In this work we present a new feature extraction method that is robust against the effects of varying vocal tract lengths. The principle of the method is based on invariant integration and makes use of a modulation filtering approach, similar to the recently proposed scattering transform. In particular, we show how the transform can be used to obtain features that are robust against variations of the vocal tract length. Phoneme recognition experiments show a clearly increased robustness in case of mismatching average vocal tract lengths.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-84890445230&origin=inward&txGid=3c31567734825773e590d808bec1b98f

U2 - 10.1109/ICASSP.2013.6639106

DO - 10.1109/ICASSP.2013.6639106

M3 - Conference contribution

T3 - IEEE International Conference on Acoustics, Speech and Signal Processing

SP - 7427

EP - 7431

BT - 2013 IEEE International Conference on Acoustics, Speech and Signal Processing

PB - IEEE

T2 - 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing

Y2 - 26 May 2013 through 31 May 2013

ER -

Feature extraction with a multiscale modulation analysis for robust automatic speech recognition

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this