Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions

Yuko Kinoshita; Takashi Osanai; Frantz Clermont

Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions

Yuko Kinoshita
, Takashi Osanai
, Frantz Clermont

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › peer-review

Abstract

This study presents the latter part of an exploratory study of the potential of sub-band parametric cepstral distance (PCD) as an alternative forensic voice comparison (FVC) feature to formants and cepstral coefficients. Using 5 Japanese vowels produced by 306 male Japanese speakers, we conducted LRbased FVC experiments under a channel mismatch condition, with sub-bands selected in reference to the expected formant locations. Combining 3 sub-band PCDs from F1, F2, and F3 ranges, sub-band PCDs outperformed the full-band PCDs in speaker classification, demonstrating their promise as an automatically extractable, robust, and linguistically interpretable acoustic feature for FVC. Index Terms: Sub-band cepstral distance, likelihood ratio, forensic voice comparison, channel mismatch, Japanese vowels

Original language	English
Title of host publication	Proceedings of the 17th Australasian International Conference on Speech Science and Technology
Editors	J Epps, J Wolfe, J Smith & C Jones
Place of Publication	Australia
Publisher	The Australasian Speech Science and Technology Association, Inc.
Pages	45-48
Edition	Peer reviewed
ISBN (Print)	2207-1296
Publication status	Published - 2018
Event	17th Australasian International Conference on Speech Science and Technology - Sydney, Australia Duration: 4 Dec 2018 → 7 Dec 2018 https://assta.org/sst-2018/ https://webarchive.nla.gov.au/awa/20190305002148/http://sst2018.unsw.edu.au/SST_proceedings/SST_2018_Proceedings_Rev_A_IDX.pdf

Conference

Conference	17th Australasian International Conference on Speech Science and Technology
Abbreviated title	SST2018
Country/Territory	Australia
City	Sydney
Period	4/12/18 → 7/12/18
Other	December 4-7 2018
Internet address	https://assta.org/sst-2018/ https://webarchive.nla.gov.au/awa/20190305002148/http://sst2018.unsw.edu.au/SST_proceedings/SST_2018_Proceedings_Rev_A_IDX.pdf

Cite this

Kinoshita, Y., Osanai, T., & Clermont, F. (2018). Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions. In J. E. J. W. J. S. . C. J. (Ed.), Proceedings of the 17th Australasian International Conference on Speech Science and Technology (Peer reviewed ed., pp. 45-48). The Australasian Speech Science and Technology Association, Inc..

Kinoshita, Yuko ; Osanai, Takashi ; Clermont, Frantz. / Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions. Proceedings of the 17th Australasian International Conference on Speech Science and Technology. editor / J Epps, J Wolfe, J Smith & C Jones. Peer reviewed. ed. Australia : The Australasian Speech Science and Technology Association, Inc., 2018. pp. 45-48

@inproceedings{4444539c9a7d48e2b3123f5e10e3be4c,

title = "Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions",

abstract = "This study presents the latter part of an exploratory study of the potential of sub-band parametric cepstral distance (PCD) as an alternative forensic voice comparison (FVC) feature to formants and cepstral coefficients. Using 5 Japanese vowels produced by 306 male Japanese speakers, we conducted LRbased FVC experiments under a channel mismatch condition, with sub-bands selected in reference to the expected formant locations. Combining 3 sub-band PCDs from F1, F2, and F3 ranges, sub-band PCDs outperformed the full-band PCDs in speaker classification, demonstrating their promise as an automatically extractable, robust, and linguistically interpretable acoustic feature for FVC. Index Terms: Sub-band cepstral distance, likelihood ratio, forensic voice comparison, channel mismatch, Japanese vowels",

author = "Yuko Kinoshita and Takashi Osanai and Frantz Clermont",

year = "2018",

language = "English",

isbn = "2207-1296",

pages = "45--48",

editor = "\{J Epps, J Wolfe, J Smith \& C Jones\}",

booktitle = "Proceedings of the 17th Australasian International Conference on Speech Science and Technology",

publisher = "The Australasian Speech Science and Technology Association, Inc.",

edition = "Peer reviewed",

note = "17th Australasian International Conference on Speech Science and Technology, SST2018 ; Conference date: 04-12-2018 Through 07-12-2018",

url = "https://assta.org/sst-2018/, https://webarchive.nla.gov.au/awa/20190305002148/http://sst2018.unsw.edu.au/SST\_proceedings/SST\_2018\_Proceedings\_Rev\_A\_IDX.pdf",

}

Kinoshita, Y, Osanai, T & Clermont, F 2018, Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions. in JEJWJSCJ (ed.), Proceedings of the 17th Australasian International Conference on Speech Science and Technology. Peer reviewed edn, The Australasian Speech Science and Technology Association, Inc., Australia, pp. 45-48, 17th Australasian International Conference on Speech Science and Technology, Sydney, New South Wales, Australia, 4/12/18.

Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions. / Kinoshita, Yuko; Osanai, Takashi; Clermont, Frantz.
Proceedings of the 17th Australasian International Conference on Speech Science and Technology. ed. / J Epps, J Wolfe, J Smith & C Jones. Peer reviewed. ed. Australia: The Australasian Speech Science and Technology Association, Inc., 2018. p. 45-48.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › peer-review

TY - GEN

T1 - Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions

AU - Kinoshita, Yuko

AU - Osanai, Takashi

AU - Clermont, Frantz

PY - 2018

Y1 - 2018

N2 - This study presents the latter part of an exploratory study of the potential of sub-band parametric cepstral distance (PCD) as an alternative forensic voice comparison (FVC) feature to formants and cepstral coefficients. Using 5 Japanese vowels produced by 306 male Japanese speakers, we conducted LRbased FVC experiments under a channel mismatch condition, with sub-bands selected in reference to the expected formant locations. Combining 3 sub-band PCDs from F1, F2, and F3 ranges, sub-band PCDs outperformed the full-band PCDs in speaker classification, demonstrating their promise as an automatically extractable, robust, and linguistically interpretable acoustic feature for FVC. Index Terms: Sub-band cepstral distance, likelihood ratio, forensic voice comparison, channel mismatch, Japanese vowels

AB - This study presents the latter part of an exploratory study of the potential of sub-band parametric cepstral distance (PCD) as an alternative forensic voice comparison (FVC) feature to formants and cepstral coefficients. Using 5 Japanese vowels produced by 306 male Japanese speakers, we conducted LRbased FVC experiments under a channel mismatch condition, with sub-bands selected in reference to the expected formant locations. Combining 3 sub-band PCDs from F1, F2, and F3 ranges, sub-band PCDs outperformed the full-band PCDs in speaker classification, demonstrating their promise as an automatically extractable, robust, and linguistically interpretable acoustic feature for FVC. Index Terms: Sub-band cepstral distance, likelihood ratio, forensic voice comparison, channel mismatch, Japanese vowels

M3 - Conference Paper

SN - 2207-1296

SP - 45

EP - 48

BT - Proceedings of the 17th Australasian International Conference on Speech Science and Technology

A2 - null, J Epps, J Wolfe, J Smith & C Jones

PB - The Australasian Speech Science and Technology Association, Inc.

CY - Australia

T2 - 17th Australasian International Conference on Speech Science and Technology

Y2 - 4 December 2018 through 7 December 2018

ER -

Kinoshita Y, Osanai T, Clermont F. Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions. In JEJWJSCJ, editor, Proceedings of the 17th Australasian International Conference on Speech Science and Technology. Peer reviewed ed. Australia: The Australasian Speech Science and Technology Association, Inc. 2018. p. 45-48

Forensic voice comparison using sub-band cepstral distances as features: A first attempt with vowels from 306 Japanese speakers under channel mismatch conditions

Abstract

Conference

Fingerprint

Cite this