The Inner Mongolia University, in the city of Hohhot, Inner Mongolia, announced on April 14 that Mongolian speech recognition and synthesis systems, the first of its kind, have been developed in China.

The systems were independently developed by the Mongolian information processing technology key lab, which is based in the School of Computing of Inner Mongolia University.

The speech recognition system uses an international deep-learning technology and is the first to offer continuous speech recognition of the large Mongolian vocabulary, with accuracy reaching more than 90 percent. The speech synthesis system uses the latest technology to ensure the speech synthesis is smooth, natural, and understandable.

The successful development of the two systems is a new technological high for Mongolian information processing, and lays a good foundation for its further improvement.

Dr. Fei Long, the main project developer, displaying the Mongolian speech recognition system with his mobile phone, on April 15. [Photo/Xinhua]

Photo taken on April 15 showing the computer interface of the Mongolian synthesis system. [Photo/Xinhua]