Our research in the field of speech technology and communication aims to understand human speech perception and production. It also deals with building computational models of speech understanding and speech generation. In order to conduct this research, the speech group analyses how humans interact verbally face-to-face and develops multimodal dialogue systems that interact via virtual or physical avatars. Thus, a common research theme is to develop human-like dialogue systems that can be used to verify theories on how humans interact when they are situated in the same physical space as well as how they handle mediated communication. Another important area is collection and analysis of conversational corpora, such as Spontal and the Stockholm Games Corpus.
The research of the Sound and Music Computing (SMC) group concerns all aspects of Sound and Music Computing, in several main streams:
The interactions between these topics are rich and diverse. An experimental approach has long been a signature of the research, often using the paradigm of analysis-by-synthesis. Additional interests include emotional expression in everyday sounds; and accessibility to music appreciation for the functionally impaired.
The robot head Furhat which was developed at TMH by Samer Al Moubayed, Gabriel Skantze, Jonas Beskow, Kalin Stefanov och Joakim Gustafson recieved the Outstanding Demo Award at The International Conference on Multimodal Interaction (ICMI).
