Analysis of speech separation for ASR systems

Author

Døbrowski, A. ; Cetnarowicz, D. ; Marciniak, T.

Author_Institution

Inst. of Control & Syst. Eng., Poznan Univ. of Technol., Poland

fYear

2004

fDate

17-20 June 2004

Firstpage

345

Lastpage

350

Abstract

This work presents some aspects concentrated on improving quality of the voice input signal for automatic speech recognition (ASR) systems. It is clear that proper separation of the desired voice from disturbances improves the recognition performance. We have tested the independent component analysis (ICA) infomax algorithm applied to the separation of isolated words. Detection of Polish words (commands used for the remote voice control of a robot) was tested using the dynamic time warping (DTW) approach.

Keywords

independent component analysis; microphone arrays; natural languages; speech processing; speech recognition; speech synthesis; time warp simulation; Polish words; automatic speech recognition system; dynamic time warping; independent component analysis; microphone array; speech separation; voice input signal; Automatic control; Automatic speech recognition; Communication system control; Independent component analysis; Microphones; Propagation delay; Signal processing; Signal processing algorithms; Speech analysis; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Robot Motion and Control, 2004. RoMoCo'04. Proceedings of the Fourth International Workshop on

Print_ISBN

83-7143-272-0

Type

conf

DOI

10.1109/ROMOCO.2004.240584

Filename

1359540