NONCAUSAL ALL-POLE MODELING OF VOICED SPEECH

This paper introduces noncausal all-pole models that are capable of ef ficiently capturing both the magnitude and phase information of voiced speech, It is shown that noncausal all-pole filter models are better able to match both magnitude and phase information and are particularl y appropriate for voiced speech due to the nature of the glottal excit ation. By modeling speech in the frequency domain, the standard diffic ulties that occur when using noncausal all-pole filters are avoided. S everal algorithms for determining the model parameters based on freque ncy-domain information and the masking effects of the ear are describe d. Our work suggests that high-quality voiced speech can be produced u sing a 14th-order noncausal all-pole model.