Third International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2003)
In this paper, we purpose a theoretical development of a metric for speech classification based on cepstral features obtained from ARMA models. Thus working with an ARMA model as a complex rational function, is possible to define a metric d(M,M´) between two stable ARMA models M, M´ by means of the cepstrum coefficients of the models. This metric may be calculated algorithmically as a finite sum in the pole-zero domain. We suggest that the metric can be used in at least two circumstances: first, we might a large number of signals that come from various types of pathological sources and we wish to classify them; alternatively, we might the underlying models Mi corresponding to several pathological voices and we wish to classify a voice (modeled as M, say) from one of those. In that case, we compute d(M,Mi) for each i and we guess the (Mi) closest to the model M.
Index Terms. ARMA model, cepstrum, distance measure, classification, pathological voice
Full Paper (reprinted with permission from Firenze University Press)
Bibliographic reference. Martínez, F. / Guillamón, A. / Martínez, J. J. (2003): "A suggested metric for cepstral ARMA-based speech classification", In MAVEBA-2003, 165-168.