The transmission part of an enhanced speech production model including the nasal tract is proposed. The discrete time system equivalent of the model is found based on the same acoustic assumptions as in the well known equivalency between a chain of tube sections and an AR-model. The transfer function is derived, but the expression is too complex to locate the singularities directly. It is possible though to determine the number of poles and zeros given the number of tube sections modelling the pharynx, the nasal tract and the upper oral tract respectively. Based on this knowledge a system identification tech- nique can be applied to the time domain implementation of the model to obtain an expression of the transfer function on normal form in which singularities are evident.
Keywords: Speech production model, nasal cavity, articulatory speech analysis, ARMA
Bibliographic reference. Olesen, Morten (1993): "Derivation of the transfer function for a speech production model including the nasal cavity", In EUROSPEECH'93, 549-552.