A comparison of DFT, PLP and cochleagram for alphabet recognition

The authors used speaker-independent classification of isolated English letters to evaluate the relative performance of the discrete Fourier transform, perceptual linear predictive (PLP) analysis, and the cochleagram auditory model. Feedforward neural network classifiers were trained using all three representations on 60 speakers and tested on 60 new speakers. Training and testing data were independently modified by adding two levels of Gaussian noise and babble (20 random letter utterances, attenuated and given random offsets). PLP gave the best results, especially when trained or tested on Gaussian noise