You are here

Dr. Khiet Truong

I am a post-doctoral researcher at the Human Media Interaction group of the University of Twente, where I am involved in the EU FP7 project SSPNet (Social Signal Processing Network). I work in the context of Social Signal Processing and Affective Computing where the goals are to develop socially and emotionally adaptive machines. My research focuses on the automatic analysis of human vocal behaviour in social conversation. I investigate human paralinguistic behaviours, for example, intonation of speech, rhythm of speech, and non-verbal vocalisations such as laughter, and relate these behaviours to social and affective meanings. The general research questions I aim to address are how do humans talk in social conversation and how can we use this knowledge to develop more socially and emotionally adaptive machines? Topics that I have been investigating in this context include feedback, turn-taking and laughing behaviour in conversation, automatic laughter detection, automatic emotion recognition in speech, recording and annotating (audiovisual) emotion corpora, and multimodal sentiment/subjectivity analysis.

My background lies in general linguistics and language and speech technology. I studied at Utrecht University and carried out my master's research at the Radboud University Nijmegen where I investigated automatic pronunciation error detection in second language learners' speech. From 2005-2009, I was employed at TNO Human Factors in Soesterberg, the Netherlands, where I worked towards my PhD on automatic emotion recognition in speech within the project MultimediaN. In 2009, I obtained my PhD in computer science from the University of Twente.

Works with