Performance evaluation in open-set speaker identification

A. Malegaonkar; A. Ariyaeeinia

doi:10.1007/978-3-642-19530-3_10

Performance evaluation in open-set speaker identification

A. Malegaonkar, A. Ariyaeeinia

Centre for Engineering Research

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

7 Citations (Scopus)

182 Downloads (Pure)

Abstract

The concern in this study is the approach to evaluating the performance of the open-set speaker identification process. In essence, such a process involves first identifying the speaker model in the database that best matches the given test utterance, and then determining if the test utterance has actually been produced by the speaker associated with the best-matched model. Whilst, conventionally, the performance of each of these two sub-processes is evaluated independently, it is argued that the use of a measure of performance for the complete process can provide a more useful basis for comparing the effectiveness of different systems. Based on this argument, an approach to assessing the performance of open-set speaker identification is considered in this paper, which is in principle similar to the method used for computing the diarisation error rate. The paper details the above approach for assessing the performance of open-set speaker identification and presents an analysis of its characteristics.

Original language	English
Title of host publication	Biometrics and ID Management. BioID 2011
Editors	C. Vielhauer, J. Dittmann, A. Drygajlo, N.C. Juul, M. C. Fairhurst
Place of Publication	Germany
Publisher	Springer Nature
Pages	106-112
Number of pages	7
Volume	6583
ISBN (Electronic)	978-3-642-19530-3
ISBN (Print)	978-3-642-19529-7
DOIs	https://doi.org/10.1007/978-3-642-19530-3_10
Publication status	Published - 10 Mar 2011
Event	European Workshop on Biometrics and Identity Management - Brandenburg an der Havel, Germany Duration: 8 Mar 2011 → 10 Mar 2011

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer, Berlin, Heidelberg

Conference

Conference	European Workshop on Biometrics and Identity Management
Abbreviated title	BioID 2011
Country/Territory	Germany
City	Brandenburg an der Havel
Period	8/03/11 → 10/03/11

Access to Document

10.1007/978-3-642-19530-3_10

905336.pdfAccepted author manuscript, 65.3 KBLicence: CC BY

Cite this

@inproceedings{3e993d1db4234479b412b772d554c122,

title = "Performance evaluation in open-set speaker identification",

abstract = "The concern in this study is the approach to evaluating the performance of the open-set speaker identification process. In essence, such a process involves first identifying the speaker model in the database that best matches the given test utterance, and then determining if the test utterance has actually been produced by the speaker associated with the best-matched model. Whilst, conventionally, the performance of each of these two sub-processes is evaluated independently, it is argued that the use of a measure of performance for the complete process can provide a more useful basis for comparing the effectiveness of different systems. Based on this argument, an approach to assessing the performance of open-set speaker identification is considered in this paper, which is in principle similar to the method used for computing the diarisation error rate. The paper details the above approach for assessing the performance of open-set speaker identification and presents an analysis of its characteristics.",

author = "A. Malegaonkar and A. Ariyaeeinia",

note = "This document is the Accepted Manuscript version of the following paper: Malegaonkar A., Ariyaeeinia A. (2011) Performance Evaluation in Open-Set Speaker Identification. In: Vielhauer C., Dittmann J., Drygajlo A., Juul N.C., Fairhurst M.C. (eds) Biometrics and ID Management. BioID 2011. Lecture Notes in Computer Science, vol 6583. Springer, Berlin, Heidelberg. The Version of Record is available online at doi: https://doi.org/10.1007/978-3-642-19530-3_10. {\textcopyright} Springer-Verlag Berlin Heidelberg 2011.; European Workshop on Biometrics and Identity Management, BioID 2011 ; Conference date: 08-03-2011 Through 10-03-2011",

year = "2011",

month = mar,

day = "10",

doi = "10.1007/978-3-642-19530-3_10",

language = "English",

isbn = "978-3-642-19529-7",

volume = "6583",

series = "Lecture Notes in Computer Science",

publisher = "Springer Nature",

pages = "106--112",

editor = "C. Vielhauer and J. Dittmann and A. Drygajlo and N.C. Juul and {Fairhurst }, {M. C. }",

booktitle = "Biometrics and ID Management. BioID 2011",

address = "Netherlands",

}

Malegaonkar, A & Ariyaeeinia, A 2011, Performance evaluation in open-set speaker identification. in C Vielhauer, J Dittmann, A Drygajlo, NC Juul & MC Fairhurst (eds), Biometrics and ID Management. BioID 2011. vol. 6583, Lecture Notes in Computer Science, Springer Nature, Germany, pp. 106-112, European Workshop on Biometrics and Identity Management, Brandenburg an der Havel, Germany, 8/03/11. https://doi.org/10.1007/978-3-642-19530-3_10

Performance evaluation in open-set speaker identification. / Malegaonkar, A.; Ariyaeeinia, A.
Biometrics and ID Management. BioID 2011. ed. / C. Vielhauer; J. Dittmann; A. Drygajlo; N.C. Juul; M. C. Fairhurst . Vol. 6583 Germany: Springer Nature, 2011. p. 106-112 (Lecture Notes in Computer Science).

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

TY - GEN

T1 - Performance evaluation in open-set speaker identification

AU - Malegaonkar, A.

AU - Ariyaeeinia, A.

N1 - This document is the Accepted Manuscript version of the following paper: Malegaonkar A., Ariyaeeinia A. (2011) Performance Evaluation in Open-Set Speaker Identification. In: Vielhauer C., Dittmann J., Drygajlo A., Juul N.C., Fairhurst M.C. (eds) Biometrics and ID Management. BioID 2011. Lecture Notes in Computer Science, vol 6583. Springer, Berlin, Heidelberg. The Version of Record is available online at doi: https://doi.org/10.1007/978-3-642-19530-3_10. © Springer-Verlag Berlin Heidelberg 2011.

PY - 2011/3/10

Y1 - 2011/3/10

N2 - The concern in this study is the approach to evaluating the performance of the open-set speaker identification process. In essence, such a process involves first identifying the speaker model in the database that best matches the given test utterance, and then determining if the test utterance has actually been produced by the speaker associated with the best-matched model. Whilst, conventionally, the performance of each of these two sub-processes is evaluated independently, it is argued that the use of a measure of performance for the complete process can provide a more useful basis for comparing the effectiveness of different systems. Based on this argument, an approach to assessing the performance of open-set speaker identification is considered in this paper, which is in principle similar to the method used for computing the diarisation error rate. The paper details the above approach for assessing the performance of open-set speaker identification and presents an analysis of its characteristics.

AB - The concern in this study is the approach to evaluating the performance of the open-set speaker identification process. In essence, such a process involves first identifying the speaker model in the database that best matches the given test utterance, and then determining if the test utterance has actually been produced by the speaker associated with the best-matched model. Whilst, conventionally, the performance of each of these two sub-processes is evaluated independently, it is argued that the use of a measure of performance for the complete process can provide a more useful basis for comparing the effectiveness of different systems. Based on this argument, an approach to assessing the performance of open-set speaker identification is considered in this paper, which is in principle similar to the method used for computing the diarisation error rate. The paper details the above approach for assessing the performance of open-set speaker identification and presents an analysis of its characteristics.

U2 - 10.1007/978-3-642-19530-3_10

DO - 10.1007/978-3-642-19530-3_10

M3 - Conference contribution

SN - 978-3-642-19529-7

VL - 6583

T3 - Lecture Notes in Computer Science

SP - 106

EP - 112

BT - Biometrics and ID Management. BioID 2011

A2 - Vielhauer, C.

A2 - Dittmann, J.

A2 - Drygajlo, A.

A2 - Juul, N.C.

A2 - Fairhurst , M. C.

PB - Springer Nature

CY - Germany

T2 - European Workshop on Biometrics and Identity Management

Y2 - 8 March 2011 through 10 March 2011

ER -