Presentation | 2006/12/14 An investigation on the speaker vector-based speaker identification with phonetic modeling Tatsuya AKATSU, Masaharu KATOH, Tetsuo KOSAKA, Masaki KOHDA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper presents a phonetic based approach for speaker identification performed in text-independent mode. The aim of this work is to improve identification performance by using information about the phonetic content of the speech. The identification systems is based on the technique of anchor models. In this system, the location of each speaker is represented by the speaker vector which consists of the set of the likelihood between a target utterance and the anchor models. In order to improve the identification performance, phonetic modeling is used instead of Gaussian Mixture Model (GMM) scheme as anchor models. This approach utilizes a phonetic speech recognizer to calculate the log-likelihood with phonetic HMMs. We also investigate the number of parameters of anchor models. The proposed method was evaluated on Japanese speaker identification task with 30 speakers. It showed that the proposed method achieved 72.1% relative improvement over the GMM-based system. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Speaker recognition / speaker identification / hidden Markov model(HMM) / Gaussian mixture model(GMM) / phonetic class |
Paper # | NLC2006-45,SP2006-101 |
Date of Issue |
Conference Information | |
Committee | NLC |
---|---|
Conference Date | 2006/12/14(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Natural Language Understanding and Models of Communication (NLC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | An investigation on the speaker vector-based speaker identification with phonetic modeling |
Sub Title (in English) | |
Keyword(1) | Speaker recognition |
Keyword(2) | speaker identification |
Keyword(3) | hidden Markov model(HMM) |
Keyword(4) | Gaussian mixture model(GMM) |
Keyword(5) | phonetic class |
1st Author's Name | Tatsuya AKATSU |
1st Author's Affiliation | Faculty of Engineering, Yamagata University() |
2nd Author's Name | Masaharu KATOH |
2nd Author's Affiliation | Faculty of Engineering, Yamagata University |
3rd Author's Name | Tetsuo KOSAKA |
3rd Author's Affiliation | Faculty of Engineering, Yamagata University |
4th Author's Name | Masaki KOHDA |
4th Author's Affiliation | Faculty of Engineering, Yamagata University |
Date | 2006/12/14 |
Paper # | NLC2006-45,SP2006-101 |
Volume (vol) | vol.106 |
Number (no) | 441 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |