Speaker classification using residence information based probabilistic model

Research Project

Project/Area Number	26350466
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Research Field	Social systems engineering/Safety system
Research Institution	National Research Institute of Police Science
Principal Investigator	KAMADA Toshiaki 科学警察研究所, 法科学第四部, 室長 (10356173)
Project Period (FY)	2014-04-01 – 2018-03-31
Project Status	Completed (Fiscal Year 2017)
Budget Amount *help	¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000) Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000) Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000) Fiscal Year 2014: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Keywords	音声情報処理 / 話者分類 / 法科学
Outline of Final Research Achievements	This study aimed to extract of residence information of speakers using speech features originating in vocal tract resonance, and local classification of speakers. We extracted the features which made residence information distance between speakers acquired by cepstrum, and speaker distribution showed the possibility of the classification by residence information of speakers.

Report

(5 results)