Microsoft researchers have created a prototype universal translation software that not only translates your speech into foreign languages but uses your own voice to output the translation. The software preserves your accent, timbre, and intonation while translating between 26 languages. The software has been developed at Microsoft’s labs in Beijing and was unveiled at TechFest 2012 by chief research officer Rick Rashid along with researcher Frank Soong.
“For a monolingual speaker traveling in a foreign country, we’ll do speech recognition followed by translation, followed by the final text to speech output [in] a different language, but still in his own voice,” explained Soong regarding one of the potential applications for the software. Beyond eliminating language barriers for travelers, the software could also assist students in learning foreign languages.
However, in its current prototype stage, it takes about one hour of training in order for the software to learn a user’s nuances of speech to create a model that can read text in the user’s voice. But once the model is tuned, the translation can happen almost in real time. Among the 26 languages currently supported by the software are English, Mandarin Chinese, Spanish, and Italian.