A Computational Model for Phonetically Responsive Spoken Dialogue Systems

This paper introduces a model for segment-level phonetic responsiveness. It is based on behavior observed in human-human interaction, and is designed to be integrated into spoken dialogue systems to capture potential phonetic variation and simulate convergence capabilities. Each step in the process is responsible for an aspect of the interaction, including monitoring the input speech and appropriately analyzing it. Various parameters can be tuned to configure the speech handling and adjust the response style. Evaluation was performed by simulating simple end-to-end dialogue scenarios, including analyzing the synthesized output of the model. The results show promising ground for further extensions.