- Creation: 01/30/2020
- Update: 02/02/2020
In 25th International Conference on Auditory Display (ICAD), 2019. Yliess Hati, Francis Rousseaux, Clement Duhart
Personal assistants are becoming more pervasive in our environments but still do not provide natural interactions. Their lack of realism in term of expressiveness and their lack of visual feedback can create frustrating experiences and make users lose patience. In this sense, we propose an end-to-end trainable neural architecture for text-driven 3D mouth animations. Previous works showed such architectures provide better realism and could open the door for integrated affective Human Computer Interaction. Our study shows that such visual feedback improves significantly users' comfort for 78\% of the candidates while slightly improving their time perception.