Utterance Verification for Text-Dependent Speaker Recognition: A Comparative Assessment Using the RedDots Corpus

Tomi Kinnunen; Md Sahidullah; Ivan Kukanov; Hector Delgado; Massimiliano Todisco; Achintya Kumar Sarkar; Nicolai Bæk Thomsen; Ville Hautamaki; Nicholas Evans; Zheng-Hua Tan

doi:10.21437/Interspeech.2016-1125

Utterance Verification for Text-Dependent Speaker Recognition: A Comparative Assessment Using the RedDots Corpus

Tomi Kinnunen, Md Sahidullah, Ivan Kukanov, Hector Delgado, Massimiliano Todisco, Achintya Kumar Sarkar, Nicolai Bæk Thomsen, Ville Hautamaki, Nicholas Evans, Zheng-Hua Tan

Department of Electronic Systems

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

32 Citations (Scopus)

Abstract

Text-dependent automatic speaker verification naturally calls for the simultaneous verification of speaker identity and spoken content. These two tasks can be achieved with automatic speaker verification (ASV) and utterance verification (UV) technologies. While both have been addressed previously in the literature, a treatment of simultaneous speaker and utterance
verification with a modern, standard database is so far lacking. This is despite the burgeoning demand for voice biometrics in a plethora of practical security applications. With the goal of improving overall verification performance, this paper reports different strategies for simultaneous ASV and UV in the context
of short-duration, text-dependent speaker verification. Experiments performed on the recently released RedDots corpus are reported for three different ASV systems and four different UV systems. Results show that the combination of utterance verification with automatic speaker verification is (almost) universally
beneficial with significant performance improvements being observed.

Original language	English
Title of host publication	Proceedings Interspeech 2016
Number of pages	5
Publisher	ISCA
Publication date	8 Sept 2016
DOIs	https://doi.org/10.21437/Interspeech.2016-1125
Publication status	Published - 8 Sept 2016
Event	Interspeech 2016 - San Francisco, CA, United States Duration: 8 Sept 2016 → 12 Sept 2016 http://www.interspeech2016.org/

Conference

Conference	Interspeech 2016
Country/Territory	United States
City	San Francisco, CA
Period	08/09/2016 → 12/09/2016
Internet address	http://www.interspeech2016.org/

Series	INTERSPEECH
ISSN	1990-9770

Access to Document

10.21437/Interspeech.2016-1125

http://www.isca-speech.org/archive/Interspeech_2016/pdfs/1125.PDF

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{a2679a5bedef479693abe72564b6c7c0,

title = "Utterance Verification for Text-Dependent Speaker Recognition: A Comparative Assessment Using the RedDots Corpus",

abstract = "Text-dependent automatic speaker verification naturally calls for the simultaneous verification of speaker identity and spoken content. These two tasks can be achieved with automatic speaker verification (ASV) and utterance verification (UV) technologies. While both have been addressed previously in the literature, a treatment of simultaneous speaker and utteranceverification with a modern, standard database is so far lacking. This is despite the burgeoning demand for voice biometrics in a plethora of practical security applications. With the goal of improving overall verification performance, this paper reports different strategies for simultaneous ASV and UV in the contextof short-duration, text-dependent speaker verification. Experiments performed on the recently released RedDots corpus are reported for three different ASV systems and four different UV systems. Results show that the combination of utterance verification with automatic speaker verification is (almost) universallybeneficial with significant performance improvements being observed.",

author = "Tomi Kinnunen and Md Sahidullah and Ivan Kukanov and Hector Delgado and Massimiliano Todisco and Sarkar, {Achintya Kumar} and Thomsen, {Nicolai B{\ae}k} and Ville Hautamaki and Nicholas Evans and Zheng-Hua Tan",

year = "2016",

month = sep,

day = "8",

doi = "10.21437/Interspeech.2016-1125",

language = "English",

series = "INTERSPEECH ",

publisher = "ISCA",

booktitle = "Proceedings Interspeech 2016",

note = "Interspeech 2016 ; Conference date: 08-09-2016 Through 12-09-2016",

url = "http://www.interspeech2016.org/",

}

Kinnunen, T, Sahidullah, M, Kukanov, I, Delgado, H, Todisco, M, Sarkar, AK, Thomsen, NB, Hautamaki, V, Evans, N & Tan, Z-H 2016, Utterance Verification for Text-Dependent Speaker Recognition: A Comparative Assessment Using the RedDots Corpus. in Proceedings Interspeech 2016. ISCA, INTERSPEECH , Interspeech 2016, San Francisco, CA, United States, 08/09/2016. https://doi.org/10.21437/Interspeech.2016-1125

TY - GEN

T1 - Utterance Verification for Text-Dependent Speaker Recognition

T2 - Interspeech 2016

AU - Kinnunen, Tomi

AU - Sahidullah, Md

AU - Kukanov, Ivan

AU - Delgado, Hector

AU - Todisco, Massimiliano

AU - Sarkar, Achintya Kumar

AU - Thomsen, Nicolai Bæk

AU - Hautamaki, Ville

AU - Evans, Nicholas

AU - Tan, Zheng-Hua

PY - 2016/9/8

Y1 - 2016/9/8

N2 - Text-dependent automatic speaker verification naturally calls for the simultaneous verification of speaker identity and spoken content. These two tasks can be achieved with automatic speaker verification (ASV) and utterance verification (UV) technologies. While both have been addressed previously in the literature, a treatment of simultaneous speaker and utteranceverification with a modern, standard database is so far lacking. This is despite the burgeoning demand for voice biometrics in a plethora of practical security applications. With the goal of improving overall verification performance, this paper reports different strategies for simultaneous ASV and UV in the contextof short-duration, text-dependent speaker verification. Experiments performed on the recently released RedDots corpus are reported for three different ASV systems and four different UV systems. Results show that the combination of utterance verification with automatic speaker verification is (almost) universallybeneficial with significant performance improvements being observed.

AB - Text-dependent automatic speaker verification naturally calls for the simultaneous verification of speaker identity and spoken content. These two tasks can be achieved with automatic speaker verification (ASV) and utterance verification (UV) technologies. While both have been addressed previously in the literature, a treatment of simultaneous speaker and utteranceverification with a modern, standard database is so far lacking. This is despite the burgeoning demand for voice biometrics in a plethora of practical security applications. With the goal of improving overall verification performance, this paper reports different strategies for simultaneous ASV and UV in the contextof short-duration, text-dependent speaker verification. Experiments performed on the recently released RedDots corpus are reported for three different ASV systems and four different UV systems. Results show that the combination of utterance verification with automatic speaker verification is (almost) universallybeneficial with significant performance improvements being observed.

U2 - 10.21437/Interspeech.2016-1125

DO - 10.21437/Interspeech.2016-1125

M3 - Article in proceeding

T3 - INTERSPEECH

BT - Proceedings Interspeech 2016

PB - ISCA

Y2 - 8 September 2016 through 12 September 2016

ER -

Utterance Verification for Text-Dependent Speaker Recognition: A Comparative Assessment Using the RedDots Corpus

Abstract

Conference

Access to Document

AUB Link

Fingerprint

Cite this