A new project from Microsoft Research lets you speak in a different language without actually having to learn how. The text-to-speech tool works much like any other in that it can take text in one language and output it as speech in another. The difference is that instead of the standard synthesized computer voice, the program outputs a voice that sounds a lot like yours. Microsoft says that the program needs around an hour of training time to learn the nuances of your voice, and right now it can translate between 26 different languages, including English, Mandarin Chinese, and Spanish. The team at Microsoft's research lab in Beijing is also working on adding speech recognition capabilities — much like the conversation mode for Google Translate — which could make the tech ideal for unilingual travellers.