Tibetannet

"Is there any computer software for Tibetan-Mandarin translation? Please recommend one, urgent need!" Internet user @rexueyongkuang0000 posted this appeal on several websites. However, since this question posed in 2014 has still not yet received an answer.

Until now Nima Tashi, director of the Modern Education Technology Center at Tibet University, says this problem is about to be solved.

"Our ‘Sunshine Tibetan-Mandarin Machine Translation System’ will be ready for users this year. When it is released, it will offer convenience for many users who require translation, increasing work efficiency by 80 percent," he says.

According to the Science and Technology Department of the Tibet Autonomous Region, in 2012 Tibet’s key scientific research project on "phrase-based Tibetan-Chinese statistical machine translation achieved landmark results—the successful application of the "Sunshine Tibetan-Chinese Machine Translation System" project. The project took nearly three years to be finishied.

Nima Tashi says that the “Sunshine Tibetan language automatic segmentation system” segments words at a rate of 1,000 Tibetan words per second, reaching a rate of 95 percent accuracy in Tibetan document drafting.

"In the technological development process, we are faced with two big challenges: Tibetan word segmentation and Tibetan-Mandarin translation. We are still constantly enriching the text collection, constantly improving the translation technology."

The Tibetan language dates back to more than 1,300 years ago. It is one of the world’s oldest written languages, and it is the basic method of communication for many Tibetans. Following the popularization of computers and information technology worldwide, and computer, mobile phone, and Internet usage in Tibet, adapting the Tibetan language to information technology has become an impending issue to be resolved.

"Information technology in the Tibetan language is still a new phenomenon of the last 20 years, from simple input methods to regulating standards, and from word processing to sentence processing and paragraph processing. It is a manifestation of both social demand and the inevitable development of technology."

For more than 20 years, Nima Tashi has led the research in Tibetan information technology. He has presided over the coding of Tibetan language according to international and national standards, developed a platform for Tibetan language computer, Tibetan language computer office kits, Tibetan mobile phones, an SMS distribution platform, and a Tibetan-Chinese bilingual long distance education system. The series of information products have filled the national and international gap in Tibetan information technology.

Yongco, head of the small research and development group said, "This project took a group of nearly 30 people almost three years to complete. And now we still need to work hard, in order to make its more accurate and to increase its learning frequency."

"The accuracy rate of this kind of machine is relatively high in the document writing. In future we also plan to improve it by adapting it to literature and religion. Our idea is that when a visitor comes to Tibet, and she encounters something interesting written in Tibetan language, she can open her phone to the translation APP software and then use it to learn the meaning of the Tibetan script before her," Yongco added.

The name given to the Tibetan language information technology computer software issued by Nima Tashi and his team is "Sunshine", for it will bring warmth and service to many people.