Microsoft Research Demonstrates Impressive Voice Translation System

 Universal Translator

At a recent conference, Microsoft Research demonstrated a speech recognition breakthrough via machine translation that converted spoken English words into computer-generated Chinese language. The breakthrough is patterned after deep neural networks and significantly reduces errors in spoken as well as written translation.

I

t has long been the intention of computer scientists and programmers to develop systems that could fluidly and automatically translate spoken words naturally and without latency.  Such a human-computer interaction model is frequently referred to as a natural user interface.

Today even on Wikipedia, the universal translator is defined as a device of science fiction. 

Ray Kurzweil is known to have worked on such applications and provided demonstrations in 2008 of semi-functional prototype systems.  In the demonstration, Kurzweil does speak in a very metered tone and tempo in the demo.  He also speaks the punctuation marks, saying 'period,' for instance.

In April AT&T also demonstrated a real-time translator system they are developing.

Now in the video below, Mircrosoft Chief Research Officer Rick Rashid demonstrates a speech recognition breakthrough via machine translation that converts his spoken English words into computer-generated Chinese language. The breakthrough is patterned after deep neural networks and significantly reduces errors in spoken as well as written translation. —

It is assured that in a few years the barrier of language will be virtually eliminated in our interactions and daily activities in our day-to-day communications.  Already for web pages, instant translation makes 33rd Squared a popular site in Russia, India, China and many other non-English speaking countries.  

Skip ahead to ~7:25 for the heart of the demonstration, however the preceding background explanation is very informative. 

SOURCE  Microsoft Research

By 33rd Square

Subscribe to 33rd Square