Next-Generation Spoken Dialogue Translation System

This proposal describes a research project towards a spoken dialog translation system that holds a broad range of potential applications. In particular, the proposal describes a system that will be developed and can be integrated in hand-held devices where a user may talk to the system in ones preferred language and the system produces a spoken translation of what is said in another language (e.g. Japanese to English or vice versa). The language barrier between any two dialog partners speaking different languages can be bridged with the help of such a system. The proposed project spans several fields of research: knowledge and expertise in automatic speech recognition (ASR), spoken language understanding (SLU), information retrieval (IR), spoken dialog design (SDD), natural language generation (NLG), speech synthesis, and mobile telecommunication technologies are required. The expected benefit to the industrial partner is to use the research output of this project in a commercial product.

Faculty Supervisor:

Dr. Gerl=ald Penn

Student:

Kinfe Tadesse Mengistu

Partner:

Soligsoft Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

University of Toronto