Výsledky vyhledávání - "Yuichiro, Mera"

Unsupervised training of phoneme models using HMM sufficient statistics and a speaker distance function

Autor: Akira Baba, Akinobu Lee, Kiyohiro Shikano, Kanako Matsunami, Shinichi Yoshizawa, Yuichiro Mera, Miichi Yamada

Publikováno v: Electronics and Communications in Japan (Part III: Fundamental Electronic Science). 88:33-41

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::84957219570c5d479a202b8ba85b9c3c
https://doi.org/10.1002/ecjc.20188

Zobrazit plný text záznamu

Evaluation on unsupervised speaker adaptation based on sufficient HMM statictics of selected speakers

Autor: Shinichi, Yoshizawa, Akira, Baba, Kanako, Matsunami, Yuichiro, Mera, Miichi, Yamada, Akinobu, Lee, Kiyohiro, Shikano

Publikováno v: 7th European Conference on Speech Communication and Technology (Eurospeech 2001).

This paper describes an efficient method of unsupervised speaker adaptation. This method is based on (1) selecting a subset of speakers who are acoustically close to a test speaker, and (2) calculating adapted model parameters according to the previo

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c97c83d04e3b4348946011d1b5b5e618
https://doi.org/10.21437/eurospeech.2001-317

Zobrazit plný text záznamu

Unsupervised Noisy Environment Adaptation Algorithm Using MLLR and Speaker Selection

Autor: Miichi Yamada, Akira Baba, Shinichi Yoshizawa, Yuichiro Mera, Akinobu Lee, Hiroshi Saruwatari, Kiyohiro Shikano

An unsupervised acoustic model adaptation algorithm using MLLR and speaker selection for noisy environments is proposed. The proposed algorithm requires only one arbitrary utterance and environmental noise data. The adaptation procedure is composed o

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c76f82823d3aaf01d8f22002cad52666
http://hdl.handle.net/10061/7961

Zobrazit plný text záznamu

Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers

Autor: Yuichiro Mera, K. Matsunami, Shinichi Yoshizawa, Akira Baba, M. Yamada, Kiyohiro Shikano

Publikováno v: ICASSP

Describes an efficient method for unsupervised speaker adaptation. This method is based on (1) selecting a subset of speakers who are acoustically close to a test speaker, and (2) calculating adapted model parameters according to the previously store

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3e93f6d44be911d0c3508c51ece0c7c6
http://hdl.handle.net/10061/8039

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání