DocumentCode :
3353943
Title :
Improvement In Automatic Speech Recognition Performance In Noisy Environments Using Time-Domain Blind Source Separation
Author :
Demir, Cemil ; Harmanci, F. Kerem
Author_Institution :
Elektrik Elektronik Muhendisligi Bolumu, Bogazici Univ., Istanbul, Turkey
fYear :
2007
fDate :
11-13 June 2007
Firstpage :
1
Lastpage :
4
Abstract :
Blind source separation (BSS) methods are generally used to separate speeches of people who are simultaneously speaking in the same room by using more than one microphone to record the speeches. However, in this study, the aim is to separate noise from the speech and therefore improve speech recognition performance using a time-domain BSS method. This method uses time-domain second-order statistics and is based on non-stationarity and non-whiteness properties of speech signals. Sphinx, which is a speaker independent, continuous speech recognition engine, is used to test the recognition performance of resulting enhanced speech. The simulation results demonstrate that, there is a remarkable improvement in recognition performance in terms of sentence error rate and word error rate.
Keywords :
blind source separation; microphones; speech recognition; time-domain analysis; Sphinx; automatic speech recognition; continuous speech recognition engine; microphone; noisy environments; sentence error rate; time domain blind source separation; time-domain second-order statistics; word error rate; Automatic speech recognition; Blind source separation; Error analysis; Microphones; Source separation; Speech enhancement; Speech recognition; Statistics; Time domain analysis; Working environment noise;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing and Communications Applications, 2007. SIU 2007. IEEE 15th
Conference_Location :
Eskisehir
Print_ISBN :
1-4244-0719-2
Electronic_ISBN :
1-4244-0720-6
Type :
conf
DOI :
10.1109/SIU.2007.4298592
Filename :
4298592
Link To Document :
بازگشت