Generalized cyclic transformations in speaker-independent speech recognition

F. Müller; E. Belilovsky; A. Mertins

doi:10.1109/ASRU.2009.5373284

Generalized cyclic transformations in speaker-independent speech recognition

F. Müller, E. Belilovsky, A. Mertins

Institute of Signal Processing

Abstract

A feature extraction method is presented that is robust against vocal tract length changes. It uses the generalized cyclic transformations primarily used within the field of pattern recognition. In matching training and testing conditions the resulting accuracies are comparable to the ones of MFCCs. However, in mismatching training and testing conditions with respect to the mean vocal tract length the presented features significantly outperform the MFCCs.

Original language	English
Title of host publication	2009 IEEE Workshop on Automatic Speech Recognition Understanding
Number of pages	5
Publisher	IEEE
Publication date	01.11.2009
Pages	211-215
ISBN (Print)	978-1-4244-5478-5
ISBN (Electronic)	978-1-4244-5479-2
DOIs	https://doi.org/10.1109/ASRU.2009.5373284
Publication status	Published - 01.11.2009
Event	2009 IEEE Workshop on Automatic Speech Recognition and Understanding - Merano, Italy Duration: 13.12.2009 → 17.12.2009 Conference number: 79490

UN SDGs

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

Access to Document

10.1109/ASRU.2009.5373284

Cite this

@inproceedings{b3e1f1f69a9446fabfd7a17e81cf7657,

title = "Generalized cyclic transformations in speaker-independent speech recognition",

abstract = "A feature extraction method is presented that is robust against vocal tract length changes. It uses the generalized cyclic transformations primarily used within the field of pattern recognition. In matching training and testing conditions the resulting accuracies are comparable to the ones of MFCCs. However, in mismatching training and testing conditions with respect to the mean vocal tract length the presented features significantly outperform the MFCCs.",

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

year = "2009",

month = nov,

day = "1",

doi = "10.1109/ASRU.2009.5373284",

language = "English",

isbn = "978-1-4244-5478-5",

series = " IEEE Workshop on Automatic Speech Recognition and Understanding",

publisher = "IEEE",

pages = "211--215",

booktitle = "2009 IEEE Workshop on Automatic Speech Recognition Understanding",

note = "2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009 ; Conference date: 13-12-2009 Through 17-12-2009",

}

Müller, F, Belilovsky, E & Mertins, A 2009, Generalized cyclic transformations in speaker-independent speech recognition. in 2009 IEEE Workshop on Automatic Speech Recognition Understanding. IEEE Workshop on Automatic Speech Recognition and Understanding, IEEE, pp. 211-215, 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, Merano, Italy, 13.12.09. https://doi.org/10.1109/ASRU.2009.5373284

Generalized cyclic transformations in speaker-independent speech recognition. / Müller, F.; Belilovsky, E.; Mertins, A.
2009 IEEE Workshop on Automatic Speech Recognition Understanding. IEEE, 2009. p. 211-215 ( IEEE Workshop on Automatic Speech Recognition and Understanding).

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

TY - GEN

T1 - Generalized cyclic transformations in speaker-independent speech recognition

AU - Müller, F.

AU - Belilovsky, E.

AU - Mertins, A.

N1 - Conference code: 79490

PY - 2009/11/1

Y1 - 2009/11/1

N2 - A feature extraction method is presented that is robust against vocal tract length changes. It uses the generalized cyclic transformations primarily used within the field of pattern recognition. In matching training and testing conditions the resulting accuracies are comparable to the ones of MFCCs. However, in mismatching training and testing conditions with respect to the mean vocal tract length the presented features significantly outperform the MFCCs.

AB - A feature extraction method is presented that is robust against vocal tract length changes. It uses the generalized cyclic transformations primarily used within the field of pattern recognition. In matching training and testing conditions the resulting accuracies are comparable to the ones of MFCCs. However, in mismatching training and testing conditions with respect to the mean vocal tract length the presented features significantly outperform the MFCCs.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-77949470277&origin=inward&txGid=639a402ad1fea2090751e60f6e737f0b

U2 - 10.1109/ASRU.2009.5373284

DO - 10.1109/ASRU.2009.5373284

M3 - Conference contribution

SN - 978-1-4244-5478-5

T3 - IEEE Workshop on Automatic Speech Recognition and Understanding

SP - 211

EP - 215

BT - 2009 IEEE Workshop on Automatic Speech Recognition Understanding

PB - IEEE

T2 - 2009 IEEE Workshop on Automatic Speech Recognition and Understanding

Y2 - 13 December 2009 through 17 December 2009

ER -

Generalized cyclic transformations in speaker-independent speech recognition

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Invariant features for automatic speech recognition

Cite this

Generalized cyclic transformations in speaker-independent speech recognition

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Projects

Invariant features for automatic speech recognition

Cite this