Intonation: Theory, Models, and Applications

Athens, Greece
September 18-20, 1997


Modelling Intonational Structure Using Hidden Markov Models

Helen Wright, Paul Taylor

Centre for Speech Technology Research, University of Edinburgh, Edinburgh, U.K.

A method is introduced for using hidden Markov models (HMMs) to model intonational structure. HMMs are probabilistic and can capture the variability in structure which previous finite state network models lack. We show how intonational tunes can be modelled by separate HMMs and how HMMs can be used in a recognition system to automatically determine the tune type of an utterance.

Full Paper

Bibliographic reference.  Wright, Helen / Taylor, Paul (1997): "Modelling intonational structure using hidden Markov models", In INT-1997, 333-336.