Auditory models with Kohonen SOFM and LVQ for speaker independent phoneme recognition

Neural networks that employed unsupervised learning were used on the output of two different models of the auditory periphery to perform phoneme recognition. Experiments which compared the performance of these two auditory model representations to mel-cepstral coefficients showed that the auditory models performed significantly better in terms of phoneme recognition accuracy under the conditions tested (high signal-to-noise and a large database of speakers). However, the three representations made different types of broad class recognition errors. The Patterson auditory model representation performed best with the highest overall phoneme and broad class performance