Personal tools

Multidisciplinary Speech and Voice Research at University of Aveiro - From Production to Interaction


Jump to: navigation, search

Title Multidisciplinary Speech and Voice Research at University of Aveiro - From Production to Interaction
Speaker António J. S. Teixeira
Institution FEUP
Date 2011/11/30
Time 13:00
Location INESC Porto
Country Portugal

Abstract "Ubiquitous computing - access to all information for anyone, anywhere, at any time - relies on speech for those whose eyes or hands are directed to other tasks (such as driving ...) or for those who cannot read or type (such as children, the blind, or the disabled)." [Nass & Brave, in Wired for Speech]

Use of speech, alone or in multimodal setups, is expected to improve our interaction with machines, more and more common in our daily tasks. Speech is our natural form of communication, allows interaction at some distance, can be used on the ubiquitous telephone, etc. But efficient use of speech for interaction challenges state-of-the-art speech technology and the available knowledge on Human-Human and Human-Machine communication. As others, we consider that “[Computer] Interaction with Humans is the key” [Victor Zue, MIT, ISCA Medalist] and have as our long term goal the development of Robust, Useful and Easy to Use Human-Machine Interaction using speech and language, supporting Portuguese native speakers. For several years, we have been doing research at both the applications and basic knowledge level, adopting an anthropomorphic approach. As von Kempelen, in 1791, we consider necessary to "have a perfect knowledge of what [we] wanted to imitate", making a "formal study of speech". In this way, our talking machines and our theories will progress, "the one serving as guide to the other". However, we do not propose that machines should be exactly like humans. After all, planes don't flap their wings to fly. What we are looking for are basic principles.

Talk will include an overview of our research at the two levels: - Applications: Focusing on the presentation of the Spoken Interface developed for an autonomous mobile robot (Carl) and ongoing work on Multimodal interaction; - Basic Research: Focusing on Speech Production studies using MRI and Articulatory Speech Synthesis (both simulation of the speech production process and application of new linguistic theories to the development of the necessary processing modules so that the articulatory synthesizer may be the base of a complete text to speech system).

Speaker António J. S. Teixeira Departamento de Electrónica Telecomunicações e Informática (DETI), Universidade de Aveiro IEETA (Signal Processing Group & Transverse Area on Intelligent Robotics & Transverse Activity on Innovative Biomedical Technologies) More on the author: