Speech Trans : An Experimental Real-Time Speech-To-Speech Translation System

Abstract

This paper reports the current progress in the SPEECHTRANS project at the Center for Machine Translation which is a speech-to-speech translation project for real-time processing of speaker-independent noisy continuous speech input. SPEECHTRANS uses a custom speech recognition hardware and a phoneme-based generalized LR parser that uses a unification-based grammar formalism and a natural language generator that is connected to a voice synthesis module. SPEECHTRANS was originally presented at the MT Conference held at CMU in June 1988 as a public demonstration of a realtime speaker-independent speech-to-speech translation system.