Additive noise and channel distortion-robust parametrization tool - performance evaluation on Aurora 2 & 3

In this paper a HTK-compatible robust speech parametrization tool CtuCopy is presented. This tool allows for the usage of several additive noise suppression preprocessing techniques, nonlinear spectrum transformation, RASTA-like filtration, and direct final feature computation. The tool is general, it is easily extendible, and it may be also used for speech enhancement purposes. In the second part, parametrizations combining the extended spectral subtraction for additive noise suppression and LDA RASTA-like filtration for channel-distortion elimination with final computation of PLP cepstral coefficients are examined and evaluated on Aurora 2 & 3 and Czech SpeechDat corpora. This comparison shows specific algorithm features and the differences in their behavior on above mentioned databases. PLP cepstral coefficients with both extended spectral subtraction and LDA RASTA-like filtration seem to be good choice for noise robust parametrization.