We describe our research direction to build a 3-D articulatory speech synthesizer. We are employing a user-centred design approach to develop the synthesizer. The end result will be added to the open-source research community. The system infrastructure we are planning is expected to allow for easy integration of current and future research results in vocal tract modelling and vocal sound produc...

With the growth of speech and language interactions in many commercially available devices and applications, it is timely that the HCI community should engage more fully with such a modality. This position paper proposes that HCI should utilize theory-based insight from dialogue research to develop in depth understanding about our behaviors in speech and language interactions with computers. Th...

Data fusion from multiple sensors is of significant interest to the speech research community, as it can potentially provide a better picture of speech production through the use of complementary sensor modalities. This paper deals with the practical aspects of this problem, such as acquisition and processing of the dynamic ultrasound (US) and electromagnetic (EM) data of the tongue during spee...

January is is the second progress report for Project /// (Generative Kernels and Score Spaces for Classiication of Speech) within the Global Uncertainties Programme. is project combines the current generative models developed in the speech community with discriminative classiiers. An important aspect of the approach is that the generative models are used to deene a score-space that can be used ...

The department is engaged in European cooperation within the COST-projec t on speech synthesis and recognition, and has also been invited to the European Community project on Blindness and Technology. Together with the Swedish Telecom, we are engaged in the SAM project (Speech Assessment Methods) in the second phase of ESPRIT. The Swedish Teleeom is an important industrial contact for the depar...

Large annotated speech corpora are a critical component of research in prosody. The classification of languages according to their speech rhythm, for example, requires a great number of annotated sentences by different speakers in different languages. We have developed Vocale, a tool for the semiautomatic annotation of vocalic and consonantal parts of speech because in recent models these units...

Several strands of research in the fields of linguistics, speech perception, and neuroethology suggest that modelling the temporal dynamics of an acoustic event landmark-based representation is a scientifically plausible approach to the automatic speech recognition (ASR) problem. Adopting a point process representation of the speech signal opens up ASR to a large class of statistical models tha...

Multimodal interfaces provide more flexible and compelling interaction and can enable public information kiosks to support more complex tasks for a broader community of users. MATCHKiosk is a multimodal interactive city guide which provides users with the freedom to interact using speech, pen, touch or multimodal inputs. The system responds by generating multimodal presentations that synchroniz...

OBJECTIVE To propose a community-based network system and interdisciplinary management for children with cleft lip/ palate, applicable to Thailand and other developing countries. MATERIAL AND METHOD A network was developed for the care of patients with cleft lip/palate by combining primary health care, community-based rehabilitation and institutional expertise (Tawanchai Center), Five worksho...

Multimodal interfaces provide more flexible and compelling interaction and can enable public information kiosks to support more complex tasks for a broader community of users. MATCHKiosk is a multimodal interactive city guide in which users are free to interact using speech, pen, touch or multimodal inputs and the system responds by generating multimodal presentations that synchronize synthetic...