Title :
Temporal and Spectral Processing of Degraded Speech
Author :
Krishnamoorthy, P. ; Prasanna, Mahadeva S R
Author_Institution :
Dept. of Electron. & Commun. Eng., Indian Inst. of Technol. Guwahati, Guwahati
Abstract :
This paper proposes the combined temporal and spectral processing approach for enhancement of degraded speech. Temporal processing refers to the processing of excitation source information in the temporal domain which mainly involves identification and enhancement of high signal-to-noise ratio (SNR) regions in the time domain representation of the degraded speech signal. Similarly spectral processing involves identification and suppression of degraded components in the frequency domain representation of the degraded speech signal.
Keywords :
signal representation; spectral analysis; speech enhancement; degraded speech signal; degraded speech suppression; excitation source information; frequency domain representation; signal-to-noise ratio; spectral processing; speech enhancement; temporal processing; Degradation; Frequency domain analysis; Nonlinear filters; Paper technology; Signal processing; Signal to noise ratio; Speech analysis; Speech enhancement; Speech processing; Working environment noise;
Conference_Titel :
Advanced Computing and Communications, 2008. ADCOM 2008. 16th International Conference on
Conference_Location :
Chennai
Print_ISBN :
978-1-4244-2962-2
Electronic_ISBN :
978-1-4244-2963-9
DOI :
10.1109/ADCOM.2008.4760436