Development of client-server speech translation system on a multi-lingual speech communication platform

This paper describes a client-server speech-to-speech translation system developed on a multi-lingual speech communication platform. This platform enables easy assembly of speech communication system from the corresponding software modules (e.g. speech recognition, spoken language machine-translation, speech synthesis). This client-server speech translation system is designed for use at mobile terminals. Terminals and servers are connected via a 3G public mobile phone networks, and speech translation services are available at various places with thin client. This system realizes hands-free communication and robustness for real use of speech translation in noisy environments. A microphone array and new noise suppression technique improves speech recognition performance, and a corpus-based approach enables wide coverage, robustness and portability to new languages and domains. Recent evaluation of the overall system showed that the utterance correctness of speech recognition output achieved 83%, and more than 88% of the utterances are correctly translated for Japanse-English and JapaneseChinese.