The analysis by synthesis of speech melody

Abstract

This paper describes the application of the analysis by synthesis paradigm to the melody of speech. Acomplete chain of processes is described from the acoustic analysis of fundamental frequency (f0), via thephonetic modelling of f0 using the Momel algorithm, to the surface phonological representation of thecurves using the INTSINT alphabet. Each step of the chain is designed as a reversible process which can beused to generate an acoustic output allowing an objective evaluation of the analysis. Finally, the currentimplementation of ProZed, a prosody editor for linguists, is described. It is argued that an explicit set ofmodelling tools like this will allow linguists to test different models of phonological structure which, it ishoped, will result in the availability of more and better data on a wide variety of languages.