Filler words as a speaker classification feature

Shunichi Ishihara; Yuko Kinoshita

Filler words as a speaker classification feature

Shunichi Ishihara, Yuko Kinoshita

Arts & Communication

Research output: A Conference proceeding or a Chapter in Book › Conference contribution

Abstract

Many of us can think of iconic words or expressions that our close friends or family often use. This study investigated the potential of idiosyncrasies in usage of filler words as a speaker classification parameter, using spontaneous Japanese speech. We estimated likelihood ratios (LRs) based on which filler words were used and how often, and performed speaker discrimination tests using the LRs. As a result we discovered that speakers’ choices of fillers can predict speaker’s identity to some degree, and that this has potential as an additional feature for speaker classification.

Original language	English
Title of host publication	Proceedings of the 13th Australasian International Conference on Speech Science and Technology
Place of Publication	Melbourne
Publisher	Australasian Speech Science and Technology Association
Pages	34-37
Number of pages	4
ISBN (Print)	9780958194631
Publication status	Published - 2010
Event	13th Australasian International Conference on Speech Science and Technology - Melbourne, Australia Duration: 14 Dec 2010 → 16 Dec 2010

Conference

Conference	13th Australasian International Conference on Speech Science and Technology
Country/Territory	Australia
City	Melbourne
Period	14/12/10 → 16/12/10

Access to Document

http://assta.org/sst/SST-10/SST2010/PDF/AUTHOR/ST100050.PDF

Cite this

@inproceedings{cbc6028afd8e40ceb26ae9493b4df501,

title = "Filler words as a speaker classification feature",

abstract = "Many of us can think of iconic words or expressions that our close friends or family often use. This study investigated the potential of idiosyncrasies in usage of filler words as a speaker classification parameter, using spontaneous Japanese speech. We estimated likelihood ratios (LRs) based on which filler words were used and how often, and performed speaker discrimination tests using the LRs. As a result we discovered that speakers{\textquoteright} choices of fillers can predict speaker{\textquoteright}s identity to some degree, and that this has potential as an additional feature for speaker classification.",

author = "Shunichi Ishihara and Yuko Kinoshita",

year = "2010",

language = "English",

isbn = "9780958194631",

pages = "34--37",

booktitle = "Proceedings of the 13th Australasian International Conference on Speech Science and Technology",

publisher = "Australasian Speech Science and Technology Association",

note = "13th Australasian International Conference on Speech Science and Technology ; Conference date: 14-12-2010 Through 16-12-2010",

}

Ishihara, S & Kinoshita, Y 2010, Filler words as a speaker classification feature. in Proceedings of the 13th Australasian International Conference on Speech Science and Technology. Australasian Speech Science and Technology Association, Melbourne, pp. 34-37, 13th Australasian International Conference on Speech Science and Technology, Melbourne, Australia, 14/12/10. <http://assta.org/sst/SST-10/SST2010/PDF/AUTHOR/ST100050.PDF>

TY - GEN

T1 - Filler words as a speaker classification feature

AU - Ishihara, Shunichi

AU - Kinoshita, Yuko

PY - 2010

Y1 - 2010

N2 - Many of us can think of iconic words or expressions that our close friends or family often use. This study investigated the potential of idiosyncrasies in usage of filler words as a speaker classification parameter, using spontaneous Japanese speech. We estimated likelihood ratios (LRs) based on which filler words were used and how often, and performed speaker discrimination tests using the LRs. As a result we discovered that speakers’ choices of fillers can predict speaker’s identity to some degree, and that this has potential as an additional feature for speaker classification.

AB - Many of us can think of iconic words or expressions that our close friends or family often use. This study investigated the potential of idiosyncrasies in usage of filler words as a speaker classification parameter, using spontaneous Japanese speech. We estimated likelihood ratios (LRs) based on which filler words were used and how often, and performed speaker discrimination tests using the LRs. As a result we discovered that speakers’ choices of fillers can predict speaker’s identity to some degree, and that this has potential as an additional feature for speaker classification.

M3 - Conference contribution

SN - 9780958194631

SP - 34

EP - 37

BT - Proceedings of the 13th Australasian International Conference on Speech Science and Technology

PB - Australasian Speech Science and Technology Association

CY - Melbourne

T2 - 13th Australasian International Conference on Speech Science and Technology

Y2 - 14 December 2010 through 16 December 2010

ER -

Filler words as a speaker classification feature

Abstract

Conference

Access to Document

Fingerprint

Cite this