Interactive Speech Translation in the Diplomat Project


    loading  Checking for direct PDF access through Ovid

Abstract

The Diplomat rapid-deployment speech-translation system is intended to allow naï ve users to communicate across a language barrier, without strong domain restrictions, despite the error-prone nature of current speech and translation technologies. In addition, it should be deployable for new languages an order of magnitude more quickly than traditional technologies. Achieving this ambitious set of goals depends in large part on allowing the users to correct recognition and translation errors interactively. We present the Multi-Engine Machine Translation (MEMT) architecture, describing how it is well suited for such an application. We then discuss our approaches to rapid-deployment speech recognition and synthesis. Finally we describe our incorporation of interactive error correction throughout the system design. We have already developed working bidirectional Croatian ⇆ English and Spanish ⇆ English systems, and have Haitian Creole ⇆ English and Korean ⇆ English versions under development.

    loading  Loading Related Articles