Lab of information technology, AI in Tibetan language to enter operation

LHASA, Nov. 20 (Xinhua) -- The key laboratory of information technology and artificial intelligence (AI) in the Tibetan language has passed its acceptance test in southwest China's Xizang Autonomous Region.

The project team has built a 1,500-hour Tibetan voice corpus, and developed an intelligent Tibetan voice auxiliary prompt system for public service organizations, as well as an intelligent Tibetan news broadcast system for mobile phones.

Team leader Nyima Tashi, from Tibet University, said the team has focused on Tibetan natural language processing and AI, the digitization of Tibetan cultural resources, and plateau big data analysis and application. It has studied Mandarin-Tibetan two-way machine translation, multilingual search engines, digitization of Tibetan ancient books, multi-scene and multi-style Tibetan font recognition and other technologies, and has built a number of Tibetan data resource libraries, said Nyima Tashi.

Integrating the above research results, the team initially developed the Tibetan language automatic recognition hardware and software system, and the intelligent all-in-one prototype machine. It has now begun to plan and gradually promote the application to typical scenarios such as education and training, media intelligent processing, government affairs informatization and medical informatization, both in Xizang and other regions.