Real-time implementation of HMM-based MMSE algorithm for speech enhancement in hearing aid applications

Author

Sheikhzadeh, H. ; Brennan, R.L. ; Sameti, H.

Author_Institution

Dept. of Electr. & Comput. Eng., Waterloo Univ., Ont., Canada

Volume

1

fYear

1995

fDate

9-12 May 1995

Firstpage

808

Abstract

We describe our recent work on the real-time implementation of a state-of-the-art HMM-based MMSE speech enhancement algorithm, where our earlier published algorithm has been approximated, optimized, and simplified. The key innovations enabling the enhancement system to run in real-time are: (1) an algorithm for automatic selection of the noise model, (2) pruning the MMSE forward calculations, (3) a new pause detection method useable for a SNR down to 0 dB, and (4) task partitioning of the entire system. A preliminary version of the real-time enhancement system is simulated on the IBM-PC 486DX2/66 and at the same time implemented on a DSP platform based on dual TMS320C30 DSP chips using single precision floating point arithmetic

Keywords

IBM computers; digital signal processing chips; error analysis; floating point arithmetic; hearing aids; hidden Markov models; medical signal processing; microcomputer applications; real-time systems; speech enhancement; DSP platform; HMM-based MMSE algorithm; IBM-PC 486DX2/66; MMSE forward calculations pruning; SNR; TMS320C30 DSP chips; automatic noise model selection algorithm; hearing aid applications; pause detection method; real-time enhancement system; real-time implementation; single precision floating point arithmetic; speech enhancement; task partitioning; Auditory system; Digital signal processing; Digital signal processing chips; Hidden Markov models; Partitioning algorithms; Real time systems; Signal processing algorithms; Speech enhancement; Wiener filter; Working environment noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on

Conference_Location

Detroit, MI

ISSN

1520-6149

Print_ISBN

0-7803-2431-5

Type

conf

DOI

10.1109/ICASSP.1995.479817

Filename

479817