Tibetan app a bridge to understanding

CHENGDU－Whenever he was away from home, Gerong, who only speaks Kamba, a dialect of Tibetan, often found himself in awkward situations because few people could understand him.

This month, however, when the 50-year-old from Garze prefecture in Sichuan province visited a hospital in the provincial capital of Chengdu, he was able to communicate smoothly with a Mandarin-speaking doctor about his illness.

Helping him conquer the language barrier was a smartphone speech recognition app called Dungkar Tibetan Keyboard, which can convert the three main Tibetan dialects－U-Tsang, Amdo, Kamba－into text and translate them into Chinese.

The app is the brainchild of a group of young engineers from iFLYTEK, an AI and speech-technology giant.

"With its new ability to recognize Tibetan dialects, the software can now meet the needs of about two-thirds of Tibetans across China," said Nyima Wangdu, product manager of the Dungkar app.

The keyboard was first launched in 2017, but only began to support speech recognition this month. It took the engineers nearly five years to enable the function after building a sufficient Kamba glossary.

"The main obstacles to developing the software were the lack of a glossary and the lack of support from Tibetan dialect linguists," Nyima Wangdu said.

To solve the problem, the group expanded the search for dialect speakers. They enlisted nearly 60 Tibetan students from Southwest Minzu University in Chengdu, who helped them compile the Kamba glossary.

On the technical side, the group set up a team dedicated to developing a new generation of versatile multilingual speech synthesis systems.

Cao Bin, one of the engineers, majored in linguistics in college. His work focused on tackling the syntactic differences between Tibetan and Chinese. For instance, he said, "March 13" is referred as "13 March" in Tibetan.

Deng Qi, a user interface designer, had to deal with aesthetic issues arising from the difference in sentence lengths of the two languages.

"While we tried to make the lengths match, we also worried about the accuracy of translations," Deng said. "We made great efforts to achieve an ideal result."

Now, the app has more than 4.8 million users in Tibetan-speaking regions such as the Tibet autonomous region and the provinces of Sichuan and Qinghai.

"I believe that language is not just a tool for communication, but also a bridge that brings people in different places closer to each other," Nyima Wangdu said.

Xinhua