Title :
Front-end post-processing using histogram equalization combined with ARMA filtering for noise robust speech recognition
Author :
Shariati, Seyedeh Saloomeh ; Ahadi, Seyed Mohammad ; Mohammadi, Karim
Author_Institution :
Dept. of Electr. Eng., Iran Univ. of Sci. & Technol., Tehran, Iran
Abstract :
In this paper, we present a new feature set for robust speech recognition based on histogram equalization (HEQ) combined with auto regressive moving average (ARMA) filtering. Cepstral vectors extracted from the clean data, modified by Mean and Variance Normalization (MVN) have been used to generate a reference histogram for histogram equalization. The proposed post-processing module also consists of ARMA temporal filtering applied to normalized cepstral coefficients. HEQ compensates for nonlinear distortions caused by noise and ARMA filtering is used for smoothing the normalized feature vectors. The results on the AURORA2 task have shown noticeable improvements in the recognition of noisy speech. The proposed front-end achieved a relative error reduction of around 60% compared to the standard Mel-Cepstral front-end.
Keywords :
autoregressive moving average processes; cepstral analysis; filtering theory; nonlinear distortion; speech recognition; ARMA temporal filtering; AURORA2 task; autoregressive moving average; cepstral vectors; front-end post-processing; histogram equalization; mean and variance normalization; mel-cepstral front-end; nonlinear distortions; normalized cepstral coefficients; post-processing module; relative error reduction; robust speech recognition; Cepstral analysis; Filtering; Histograms; Noise; Robustness; Speech; Speech recognition;
Conference_Titel :
Signal Processing Conference, 2007 15th European
Conference_Location :
Poznan
Print_ISBN :
978-839-2134-04-6