Section: Application Domains

Application Domains

Our research is applied in a variety of fields from ASR to paramedical domains. Speech analysis methods will contribute to the development of new technologies for language learning (for hearing-impaired persons and for the teaching of foreign languages) as well as for hearing aids. In the past, we developed a set of teaching tools based on speech analysis and recognition algorithms of the group (cf. the ISAEUS [88] project of the EU that ended in 2000). We are continuing this effort towards the diffusion of a course on Internet.

Speech is likely to play an increasing role in man-machine communication. Actually, speech is a natural mean of communication, particularly for non-specialist persons. In a multimodal environment, the association of speech and designation gestures on touch screens can, for instance, simplify the interpretation of spatial reference expressions. Besides, the use of speech is mandatory in many situations where a keyboard is not available: mobile and on-board applications (for instance in the framework of the HIWIRE European project for the use of speech recognition in a cockpit plane), interactive vocal servers, telephone and domestic applications, etc. Most of these applications will necessitate to integrate the type of speech understanding process that our group is presently studying. Furthermore, speech to speech translation concerns all multilingual applications (vocal services, audio indexing of international documents). The automatic indexing of audio and video documents is a very active field that will have an increasing importance in our group in the forthcoming years, with applications such as economic intelligence, keyword spotting and automatic categorization of mails.