Ivan Tashev, Andrew Lovitt, and Alex Acero
24 August 2009
In this paper we describe a generic architecture for single channel speech enhancement. We assume processing in frequency domain and suppression based speech enhancement methods. The framework consists of a two stage voice activity detector, noise variance estimator, a suppression rule, and an uncertain presence of the speech signal modifier. The evaluation corpus is a synthetic mixture of a clean speech (TIMIT database) and in-car recorded noises. Using the framework multiple speech enhancement algorithms are tuned for maximum performance. We propose a formalized procedure for automated tuning of these algorithms. The optimization criterion is a weighted sum of the mean opinion score (PESQ-MOS), signalto-noise-ratio (SNR), log-spectral distance (LSD), and mean square error (MSE). The proposed framework provides a complete speech enhancement chain and can be used for evaluation and tuning of other suppression rules and voice activity detector algorithms.
|Published in||2009 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing|
|Address||Victoria B.C., Canada|
© 2008 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE. http://www.ieee.org/