Paper: Personalising Speech-To-Speech Translation in the EMIME Project

ACL ID P10-4009
Title Personalising Speech-To-Speech Translation in the EMIME Project
Venue Annual Meeting of the Association of Computational Linguistics
Session System Demonstration
Year 2010
Authors

In the EMIME project we have studied un- supervised cross-lingual speaker adapta- tion. We have employed an HMM statisti- cal framework for both speech recognition and synthesis which provides transfor- mation mechanisms to adapt the synthe- sized voice in TTS (text-to-speech) using the recognized voice in ASR (automatic speech recognition). An important ap- plication for this research is personalised speech-to-speech translation that will use the voice of the speaker in the input lan- guage to utter the translated sentences in the output language. In mobile environ- ments this enhances the users’ interaction across language barriers by making the output speech sound more like the origi- nal speaker’s way of speaking, even if she or he could not speak the output language.