Comparison of auto-regressive, non-stationary excited signal parameter estimation methods

Akira Sasou; Masataka Goto; Satoru Hayamizu; Kazuyo Tanaka

Comparison of auto-regressive, non-stationary excited signal parameter estimation methods

Akira Sasou, Masataka Goto, Satoru Hayamizu, Kazuyo Tanaka

研究成果: Conference contribution

抄録

Previously, we proposed an Auto-Regressive Hidden Markov Model (AR-HMM) and an accompanying parameter estimation method. An AR-HMM was obtained by combining an AR process with an HMM introduced as a non-stationary excitation model. We demonstrated that the AR-HMM can accurately estimate the characteristics of both articulatory systems and excitation signals from high-pitched speech. As the parameter estimation method iteratively executes learning processes of HMM parameters, the proposed method was calculation-intensive. Here, we propose two novel kinds of auto-regressive, non-stationary excited signal parameter estimation methods to reduce the amount of calculation required.

本文言語	English
ホスト出版物のタイトル	Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop
編集者	A. Barros, J. Principe, J. Larsen, T. Adali, S. Douglas
ページ	295-304
ページ数	10
出版ステータス	Published - 2004
外部発表	はい
イベント	Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop - Sao Luis, Brazil 継続期間: 2004 9月 29 → 2004 10月 1

出版物シリーズ

名前	Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop

Conference

Conference	Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop
国/地域	Brazil
City	Sao Luis
Period	04/9/29 → 04/10/1

ASJC Scopus subject areas

工学（全般）

他のファイルとリンク

引用スタイル

Sasou, A., Goto, M., Hayamizu, S., & Tanaka, K. (2004). Comparison of auto-regressive, non-stationary excited signal parameter estimation methods. In A. Barros, J. Principe, J. Larsen, T. Adali, & S. Douglas (Eds.), Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop (pp. 295-304). (Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop).

Comparison of auto-regressive, non-stationary excited signal parameter estimation methods. / Sasou, Akira; Goto, Masataka; Hayamizu, Satoru その他.
Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop. ed. / A. Barros; J. Principe; J. Larsen; T. Adali; S. Douglas. 2004. p. 295-304 (Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop).

研究成果: Conference contribution

Sasou, A, Goto, M, Hayamizu, S & Tanaka, K 2004, Comparison of auto-regressive, non-stationary excited signal parameter estimation methods. in A Barros, J Principe, J Larsen, T Adali & S Douglas (eds), Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop. Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop, pp. 295-304, Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop, Sao Luis, Brazil, 04/9/29.

Sasou A, Goto M, Hayamizu S, Tanaka K. Comparison of auto-regressive, non-stationary excited signal parameter estimation methods. In Barros A, Principe J, Larsen J, Adali T, Douglas S, editors, Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop. 2004. p. 295-304. (Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop).

Sasou, Akira ; Goto, Masataka ; Hayamizu, Satoru その他. / Comparison of auto-regressive, non-stationary excited signal parameter estimation methods. Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop. editor / A. Barros ; J. Principe ; J. Larsen ; T. Adali ; S. Douglas. 2004. pp. 295-304 (Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop).

@inproceedings{cbdb985fe4154e0280d00fd8aa06b3f3,

title = "Comparison of auto-regressive, non-stationary excited signal parameter estimation methods",

abstract = "Previously, we proposed an Auto-Regressive Hidden Markov Model (AR-HMM) and an accompanying parameter estimation method. An AR-HMM was obtained by combining an AR process with an HMM introduced as a non-stationary excitation model. We demonstrated that the AR-HMM can accurately estimate the characteristics of both articulatory systems and excitation signals from high-pitched speech. As the parameter estimation method iteratively executes learning processes of HMM parameters, the proposed method was calculation-intensive. Here, we propose two novel kinds of auto-regressive, non-stationary excited signal parameter estimation methods to reduce the amount of calculation required.",

author = "Akira Sasou and Masataka Goto and Satoru Hayamizu and Kazuyo Tanaka",

year = "2004",

language = "English",

isbn = "0780386086",

series = "Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop",

pages = "295--304",

editor = "A. Barros and J. Principe and J. Larsen and T. Adali and S. Douglas",

booktitle = "Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop",

note = "Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop ; Conference date: 29-09-2004 Through 01-10-2004",

}

TY - GEN

T1 - Comparison of auto-regressive, non-stationary excited signal parameter estimation methods

AU - Sasou, Akira

AU - Goto, Masataka

AU - Hayamizu, Satoru

AU - Tanaka, Kazuyo

PY - 2004

Y1 - 2004

N2 - Previously, we proposed an Auto-Regressive Hidden Markov Model (AR-HMM) and an accompanying parameter estimation method. An AR-HMM was obtained by combining an AR process with an HMM introduced as a non-stationary excitation model. We demonstrated that the AR-HMM can accurately estimate the characteristics of both articulatory systems and excitation signals from high-pitched speech. As the parameter estimation method iteratively executes learning processes of HMM parameters, the proposed method was calculation-intensive. Here, we propose two novel kinds of auto-regressive, non-stationary excited signal parameter estimation methods to reduce the amount of calculation required.

AB - Previously, we proposed an Auto-Regressive Hidden Markov Model (AR-HMM) and an accompanying parameter estimation method. An AR-HMM was obtained by combining an AR process with an HMM introduced as a non-stationary excitation model. We demonstrated that the AR-HMM can accurately estimate the characteristics of both articulatory systems and excitation signals from high-pitched speech. As the parameter estimation method iteratively executes learning processes of HMM parameters, the proposed method was calculation-intensive. Here, we propose two novel kinds of auto-regressive, non-stationary excited signal parameter estimation methods to reduce the amount of calculation required.

UR - http://www.scopus.com/inward/record.url?scp=17644375209&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17644375209&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:17644375209

SN - 0780386086

SN - 9780780386082

T3 - Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop

SP - 295

EP - 304

BT - Machine Learning for Signal Processing XIV - Proceedings of 2004 IEEE Signal Processing Society Workshop

A2 - Barros, A.

A2 - Principe, J.

A2 - Larsen, J.

A2 - Adali, T.

A2 - Douglas, S.

T2 - Machine Learning for Signal Processing XIV - Proceedings of the 2004 IEEE Signal Processing Society Workshop

Y2 - 29 September 2004 through 1 October 2004

ER -

Comparison of auto-regressive, non-stationary excited signal parameter estimation methods

抄録

出版物シリーズ

Conference

ASJC Scopus subject areas

他のファイルとリンク

フィンガープリント

引用スタイル