The Speech group, headed by Prof. Björn Granström, is a well-known and internationally distinguished research group internationally, with very strong publication record. This standing is evidenced by its wide participation in EU projects and that it represents KTH on the boards of the International Speech Communication Association (ISCA) and the European Language and Speech Network (ELSNET). The group is currently involved in half-a-dozen national and international research projects within our current research focus on interactional aspects of spoken face-to-face communication. The group's research areas cover basic research in speech analysis, synthesis and recognition, and in language technology areas such as multimodal spoken dialogue systems and communication aids for persons with disabilities. The aim is to understand spoken human-human interaction and how this knowledge can be used in human-machine interaction. Work on multimodal dialogue systems combines research in speech technology with linguistics, phonetics, cognitive science, psychology, and computer science. The department is active in organizing conferences, summer schools and workshops. The Centre for Speech Technology (CTT) at the department is a platform for co-operation between industry and academy within the strategic area of speech technology. Generously supported by Vinnova and participating companies and organisations for ten years, CTT now continues with project funding in cooperation with its partners.
The project aims at developing a computer program that will be used to train Norwegian as a second language. [more]
GetHomeSafe - Extended Multimodal Search and Communication Systems for Safe In-Car Application Funding: EU
The aim of the proposed project is to develop a system for safe information access and communication while driving. [more]
IURO - Interactive Urban Robot Funding: EU
The goal of IURO project is to develop a robot that can engage in information-gathering face-to-face interactions in multi-user settings. [more]
Lipread - Lipreading to support Foreign and Second Language Learning by Hearing Impaired and Normally Hearing Persons Funding: EU
The objective of the project is to develop a syllabus and an e-learning environment to teach lipreading of foreign and second languages. Our target groups: deaf and hard of hearing adults, their therapists and teachers, and normally hearing language learners.
[more]
MASSIVE - Large-scale massively multimodal modelling of non-verbal behaviour in spontaneous dialogue Funding: VR
The aim is to provide a large-scale kinematic database based on motion capture of human conversational behaviour, as well as to build statistical models of multimodal non-verbal behaviour in dialogue. [more]
SAMINK - Incremental processing in multimodal conversational systems Funding: VR
The aim of this project is to understand how conversational systems can engage in spoken face-to-face interaction in real-time. [more]
SAMPAUS - Classifying and deploying pauses for flow control in conversational systems Funding: VR
The project investigates how dialog system can employ pauses and fillers to help users better understand system utterance structure. [more]
SAMPROS - Prosody in conversation Funding: RJ (Bank of Sweden Tercentenary Foundation)
The project investigates how people talking to each other jointly decide who should speak when, and the role of prosody in making these joint decisions. [more]
SAMRYTM - The rhythm of conversation Funding: VR
The project Rhythm of conversation investigates how a set of rhythmic prosodic features contributes to the joint interaction control in conversations. [more]
SAMSYNT - Introducing interactional phenomena in speech synthesis Funding: VR
The project will develop and verify ways of including interactional phenomena in speech synthesis, resulting in well-described and tested methods for synthesizing these phenomena in such a way that they can be employed to recreate human interactional behaviour. [more]
SAVIR - Situated Audio Visual Interaction with Robots Funding: SRA/KTH
The projects investigate how a robot can improve its visual scene understanding by engaging in spoken dialogue with a human.
[more]
TIVOLI - Sign learning via game-based interaction Funding: PTS - Post och Telestyrelsen
TIVOLI aims to create learning application for sign language signs, in the form of a computer game featuring sign recogntion via webcam and a signing avatar. The target group is children with communication disorders. [more]
VariQ - Intonational variation in questions in Swedish Funding: VR
This project investigates questions in dialogue. What is a question, and what makes it into one? [more]
