Evaluation of the SPLICE algorithm on the Aurora2 database

This paper describes recent improvements to SPLICE, Stereo-based Piecewise Linear Compensation for Environments, which produces an estimate of cepstrum of undistorted speech given the observed cepstrum of distorted speech. For distributed speech recognition applications, SPLICE can be placed at the server, thus limiting the processing that would take place at the client. We evaluated this algorithm on the Aurora2 task, which consists of digit sequences within the TIDigits database that have been digitally corrupted by passing them through a linear filter and/or by adding different types of realistic noises at SNRs ranging from 20dB to -5dB. For clean acoustic models, we achieve a 67.39% average decrease in word error rate over all test sets. For retrained multi-style acoustic models, the average decrease is 27.87%. The average relative word error rate reduction is 47.63%.