Mobile phone identification from speech recordings using Weighted Support Vector Machine

Yuechi Jiang; Hung Fat Frank Leung

doi:10.1109/IECON.2016.7793279

Mobile phone identification from speech recordings using Weighted Support Vector Machine

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

15 Citations (Scopus)

Abstract

In this paper, we propose a mobile phone identifier called Weighted Support Vector Machine with Weighted Majority Voting (WSVM-WMV) for a closed-set mobile phone identification task. The proposed WSVM-WMV can be regarded as a generalization of the traditional SVM identifier. On using Mel-frequency Cepstral Coefficients (MFCC) and Linear-frequency Cepstral Coefficients (LFCC) as the feature vectors, the proposed identifier can improve the identification accuracy from 92.42% to 97.86% and from 90.44% to 98.33% respectively, as compared with the traditional SVM identifier in identifying a set of 21 mobile phones.

Original language	English
Title of host publication	Proceedings of the IECON 2016 - 42nd Annual Conference of the Industrial Electronics Society
Publisher	IEEE Computer Society
Pages	963-968
Number of pages	6
ISBN (Electronic)	9781509034741
DOIs	https://doi.org/10.1109/IECON.2016.7793279
Publication status	Published - 21 Dec 2016
Event	42nd Conference of the Industrial Electronics Society, IECON 2016 - Palazzo dei Congressi, Florence, Italy Duration: 24 Oct 2016 → 27 Oct 2016

Conference

Conference	42nd Conference of the Industrial Electronics Society, IECON 2016
Country/Territory	Italy
City	Florence
Period	24/10/16 → 27/10/16

Keywords

Audio forensics
Mobile phone identification
Weighted Majority Voting
Weighted Support Vector Machine

ASJC Scopus subject areas

Control and Systems Engineering
Electrical and Electronic Engineering

More information

10.1109/IECON.2016.7793279

http://hdl.handle.net/10397/107253

Cite this

@inproceedings{58b1c7577e5b4964af620044e8188bd0,

title = "Mobile phone identification from speech recordings using Weighted Support Vector Machine",

abstract = "In this paper, we propose a mobile phone identifier called Weighted Support Vector Machine with Weighted Majority Voting (WSVM-WMV) for a closed-set mobile phone identification task. The proposed WSVM-WMV can be regarded as a generalization of the traditional SVM identifier. On using Mel-frequency Cepstral Coefficients (MFCC) and Linear-frequency Cepstral Coefficients (LFCC) as the feature vectors, the proposed identifier can improve the identification accuracy from 92.42% to 97.86% and from 90.44% to 98.33% respectively, as compared with the traditional SVM identifier in identifying a set of 21 mobile phones.",

keywords = "Audio forensics, Mobile phone identification, Weighted Majority Voting, Weighted Support Vector Machine",

author = "Yuechi Jiang and Leung, {Hung Fat Frank}",

year = "2016",

month = dec,

day = "21",

doi = "10.1109/IECON.2016.7793279",

language = "English",

pages = "963--968",

booktitle = "Proceedings of the IECON 2016 - 42nd Annual Conference of the Industrial Electronics Society",

publisher = "IEEE Computer Society",

address = "United States",

note = "42nd Conference of the Industrial Electronics Society, IECON 2016 ; Conference date: 24-10-2016 Through 27-10-2016",

}

Jiang, Y & Leung, HFF 2016, Mobile phone identification from speech recordings using Weighted Support Vector Machine. in Proceedings of the IECON 2016 - 42nd Annual Conference of the Industrial Electronics Society., 7793279, IEEE Computer Society, pp. 963-968, 42nd Conference of the Industrial Electronics Society, IECON 2016, Florence, Italy, 24/10/16. https://doi.org/10.1109/IECON.2016.7793279

Mobile phone identification from speech recordings using Weighted Support Vector Machine. / Jiang, Yuechi; Leung, Hung Fat Frank.
Proceedings of the IECON 2016 - 42nd Annual Conference of the Industrial Electronics Society. IEEE Computer Society, 2016. p. 963-968 7793279.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Mobile phone identification from speech recordings using Weighted Support Vector Machine

AU - Jiang, Yuechi

AU - Leung, Hung Fat Frank

PY - 2016/12/21

Y1 - 2016/12/21

N2 - In this paper, we propose a mobile phone identifier called Weighted Support Vector Machine with Weighted Majority Voting (WSVM-WMV) for a closed-set mobile phone identification task. The proposed WSVM-WMV can be regarded as a generalization of the traditional SVM identifier. On using Mel-frequency Cepstral Coefficients (MFCC) and Linear-frequency Cepstral Coefficients (LFCC) as the feature vectors, the proposed identifier can improve the identification accuracy from 92.42% to 97.86% and from 90.44% to 98.33% respectively, as compared with the traditional SVM identifier in identifying a set of 21 mobile phones.

AB - In this paper, we propose a mobile phone identifier called Weighted Support Vector Machine with Weighted Majority Voting (WSVM-WMV) for a closed-set mobile phone identification task. The proposed WSVM-WMV can be regarded as a generalization of the traditional SVM identifier. On using Mel-frequency Cepstral Coefficients (MFCC) and Linear-frequency Cepstral Coefficients (LFCC) as the feature vectors, the proposed identifier can improve the identification accuracy from 92.42% to 97.86% and from 90.44% to 98.33% respectively, as compared with the traditional SVM identifier in identifying a set of 21 mobile phones.

KW - Audio forensics

KW - Mobile phone identification

KW - Weighted Majority Voting

KW - Weighted Support Vector Machine

UR - http://www.scopus.com/inward/record.url?scp=85010028229&partnerID=8YFLogxK

U2 - 10.1109/IECON.2016.7793279

DO - 10.1109/IECON.2016.7793279

M3 - Conference article published in proceeding or book

SP - 963

EP - 968

BT - Proceedings of the IECON 2016 - 42nd Annual Conference of the Industrial Electronics Society

PB - IEEE Computer Society

T2 - 42nd Conference of the Industrial Electronics Society, IECON 2016

Y2 - 24 October 2016 through 27 October 2016

ER -

Mobile phone identification from speech recordings using Weighted Support Vector Machine

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this