Recursive estimation based on the trended hidden Markov model in speech enhancement

Author

Lee, Ki Yong ; Rheem, Jae Yeol ; Shirai, K.

Author_Institution

Dept. of Electron. Eng., Changwon Nat. Univ., Kyengnam, South Korea

fYear

1996

fDate

18-21 Nov 1996

Firstpage

239

Lastpage

242

Abstract

In this study, we propose a new speech enhancement based on the trended HMM. The trended HMM is a nonstationary state HMM for modeling nonstationary speech. The proposed method is a recursive method based on frame-by-frame using a Kalman filter with LP parameters controlled by a Markov switching sequence. Experimental results clearly show the superior performance of the proposed method over the standard HMM based method

Keywords

Kalman filters; hidden Markov models; recursive estimation; speech enhancement; Kalman filter; LP parameters; Markov switching sequence; frame-by-frame technique; nonstationary speech; nonstationary state HMM; recursive estimation; speech enhancement; trended HMM; trended hidden Markov model; Acoustic noise; Computer science education; Educational technology; Hidden Markov models; Recursive estimation; Speech analysis; Speech enhancement; Speech processing; State estimation; White noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1996., IEEE Asia Pacific Conference on

Conference_Location

Seoul

Print_ISBN

0-7803-3702-6

Type

conf

DOI

10.1109/APCAS.1996.569263

Filename

569263

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2180193