Microsoft Research has developed a system to not only translate spoken English into Mandarin, but also to output the Mandarin speech in the vocal style of the original speaker in a matter of seconds. As demonstrated in the video above, English sentences can be spoken into the new speech-recognition, translation and generation system and the words will be output into Mandarin while preserving the speaker’s tone and vocal cadence. The preservation of voice, can aid in conveying the meaning of the spoken sentences. Once perfected, this technology could be very useful for health professionals who need to communicate with patients who speak different languages.
(Source: New Scientist)