Compensating Acoustic Mismatch Using Class-Based Histogram Equalization for Robust Speech Recognition. (22nd March 2007)

• Record Type: Journal Article
• Authors: Suh Suh, Youngjoo Youngjoo; Kim Kim, Sungtak Sungtak; Kim Kim, Hoirin Hoirin
• Other Names: Gales Gales Mark Mark Academic Editor.
• Is Part Of: EURASIP journal on advances in signal processing. Volume 2007(2007)
• Publication Date: 2007-03-22
• Subjects: Signal processing -- Periodicals; Traitement du signal; Signal processing; Periodicals; 621.3822
• DOI: 10.1155/2007/67870
• Languages: English
• ISSNs: 1687-6172

Abstract

Abstract : A new class-based histogram equalization method is proposed for robust speech recognition. The proposed method aims at not only compensating for an acoustic mismatch between training and test environments but also reducing the two fundamental limitations of the conventional histogram equalization method, the discrepancy between the phonetic distributions of training and test speech data, and the nonmonotonic transformation caused by the acoustic mismatch. The algorithm employs multiple class-specific reference and test cumulative distribution functions, classifies noisy test features into their corresponding classes, and equalizes the features by using their corresponding class reference and test distributions. The minimum mean-square error log-spectral amplitude (MMSE-LSA)-based speech enhancement is added just prior to the baseline feature extraction to reduce the corruption by additive noise. The experiments on the Aurora2 database proved the effectiveness of the proposed method by reducing relative errors by62 % over the mel-cepstral-based features and by23 % over the conventional histogram equalization method, respectively.