Spoken Language Understanding for human-computer interaction

Starting: 01/10/2015
PhD Student: Edwin Simonnet
Advisor(s): Yannick Estève (LIUM, LST)
Co-advisor(s): Nathalie Camelin (LIUM, LST)
Funding: Projet Européen (JOKER) / Région

This thesis work is done in connection with the European project JOKER (JOKe and Empathy of a Robot/ECA: Towards social and affective relations with a robot). This project is funded by the program CHIST-ERA (European Coordinated Research on Long-term Challenges in Information and Communication Sciences & Technologies ERA-Net).

The JOKER project aims to develop new approaches for social and emotional interaction between a human and a companion robot. This thesis subject requires from the student good skills in the field of deep learning, speech recognition, and automatic language processing for semantic information extraction. It is also necessary to develop high-level software tools, put in place rigorous experimental protocols, and apply them.

The different implementations of neural architectures use Python and the Theano library. It is possible that in the context of the thesis the automatic speech recognition system will evolve to a solution more conducive to the task of spoken language understanding.