TIDE PROJECT TP1174 VAESS

- Project presentation

General Project Overview

Project Number	TP1174
Sector	TIDE
Title	VOICES, ATTITUDES AND EMOTIONS IN SPEECH SYNTHESIS
Acronym	VAESS

Project Summary

AIMS

The primary aim of the VAESS partners is to develop improved quality and range of synthetic voices in a number of languages and on a number of synthesisers including a small and portable, yet powerful, communication aid. The provision of new voices will be simplified and (at least partially) automated. The methods developed will be tested by implementing male, female, and if time permits, children's voices. An attempt will be made to include a range of attitudes and emotions in the synthesised speech, and to provide efficient user control of these features. In view of the expertise and facilities available to the partners, these factors will be studied solely for Danish and Swedish. The proposed communication aid will be based on PC-compatible hardware, and thus the algorithms developed will function on other standard PC platforms.

APPROACH

To ensure that the final system is useful to (and usable by) the disabled and the elderly, it is vital that the system will be simple and easy for computer-illiterate people to understand. This must include both the "training phase" (where the user and their carer will work together to develop a new personalised voice) and the "operating phase" (where the user will operate the system alone, for day-to-day communication). Consequently, an important part of the project will be the design of the user interface. Similarly, it is important that the automatic methods used to analyse speech are reliable and do not need specialist intervention. Overall, the structure of the project will include the following steps: 1) Study and manual labelling of speech features for voices, attitude and emotion. 2) Development of automatic or semi-automatic labelling of above and conversion into text-to-speech control parameters. 3) Hardware platform development for operation with and without DSP. 4) Development of the user interface. 5) Integration of the systems. 6) Evaluation of the systems and completion of the user interface.

IMPACT

Currently the use of speech synthesis based on communication aids is adversely affected by their inappropriate and artificial voices, and by limited ability to express emotion efficiently. Provision of a genuinely personalised voice would be a major advantage over current aids. The present Infovox synthesis technology, available for some ten European languages, will be augmented and generalised throughout the project, to enable use in a variety of present and future devices for the elderly and disabled people. The combination of the Infovox speech technology and the small portable BiDesign platform will offer very strong competition to current European and non-European systems alike. Although the initial aim of the project is to produce a flexible speech communication aid for disabled people, the techniques developed would also have significant potential for exploitation in devices for non-disabled users. In particular, multi-media and home entertainment systems would benefit from more varied vocal styles, or even ones which could be modified by the users. Other potential aids for disabled people would include "talking newspapers" and electronic messaging systems (E-mail) with voices customised by the user to increase intelligibility and reduce "listening fatigue".

EXPLOITATION

The wealth of experience in Speech research and rehabilitation applications of communication aids from the four university partners and commercial expertise of the two industrial partners ensures optimal opportunity to succeed. One university and one industrial partner come from Sweden while the remaining partners come from three EC countries. All the deliverables from this project have the potential for commercial exploitation. However, the easiest to market would be the synthesis system as a whole, and the hardware platform on which it runs. The availability of such systems for two of the most widely-used languages worldwide (English and Spanish) should make subsequent exploitation highly profitable. The commercial partners and the universities will coordinate their efforts to ensure that the results of the project are successfully exploited by making use of their existing industrial and marketing contacts. The results of research and fundamental theoretical issues will be disseminated through EC and other international conferences and publications.