REPORT Na_Idiap-RR-05-2013/IDIAP Convolutional Pitch Target Approximation Model for Speech Synthesis Na, Xingyu Garner, Philip N. Pitch modelling pitch target approximation speech synthesis EXTERNAL https://publications.idiap.ch/attachments/reports/2013/Na_Idiap-RR-05-2013.pdf PUBLIC Idiap-RR-05-2013 2013 Idiap March 2013 In this paper, we investigate pitch contour modelling in speech synthesis based on segmental units. A convolutional pitch target approximation model is proposed. This model allows jointly stochastic modelling of framewise pitch and pitch contour of longer units, of which the intuitive relations are revealed by a convolutional target approximation filter. The pitch contour is stylized by a linear representation called pitch target. In synthesis stage, the likelihood of the framewise model and the pitch target model are jointly maximized using a Toeplitz matrix representing the discrete convolutional filter.