Joakim Gustafson

Joakim Gustafson is a Professor of Speech Technology and Head of the Department at the Division of Speech, Music and Hearing, KTH Royal Institute of Technology. His research focuses on multimodal systems, conversational speech synthesis, social robotics, and interactional analysis of spontaneous spoken dialogue. He leads VR-funded projects such as CONNECTED (context-aware speech synthesis for conversational AI) and co-leads projects like STANCE (speaker stance perception) and CAPTivating (public speaking analysis with TTS). He is also involved in Digital Futures-funded AAIS (social robots for elderly assistance) and WASP-funded PerCorSo (robot behavior in crowded environments).

Education: Defended his PhD thesis at KTH in 2002 under supervision of Björn Granström, with faculty opponent Julia Hirschberg from Columbia University. Prior roles include Senior Researcher at Telia Research (2000-2007), leading projects like NICE (speech-enabled computer game for children) and TänkOm (animated agent Pixie).

Research interests span speech synthesis realism, robot interaction design, and accessibility technologies. Projects aim to improve AAC devices, dementia detection via multimodal analysis, and sustainable cooking interfaces. He actively contributes to editorial boards of journals including Speech Communication and International Journal of Human-Computer Studies, and served as Technical Program co-Chair for Interspeech 2017. His lab, the Intelligence Augmentation Lab, supports data collection for projects like Food Talk.

Grants & Collaborations: VR, RJ, WASP, Vinnova, and EU funding across multiple projects. Co-PI in international initiatives like BabyRobot (robot learning for children) and EACare (dementia early detection).

Labs/Teams: Intelligence Augmentation Lab (IA-Lab) at KTH, collaborating with RPL (Robot Perception & Language Lab) and ISCA (International Speech Communication Association).