Speech synthesis and recognition in robotic systems designed to help people

Human auditory characteristics make speech a primary modality in human-machine interaction. In robotics, accurate recognition of critical situations and informing it to an operator can be especially useful to prevent any potential dangers and to achieve the main objectives of the application.

In this project, human-like natural speech synthesis, highly robust speech recognition, multimodal interaction, situation recognition, and complex dialogue management methods will be developed. The system suggested in this project can potentially be very useful for people with a speech impediment or children having limited language skills.