LHASA, June 22 (Xinhua) -- In the laboratory at Xizang University in southwest China's Xizang Autonomous Region, Yong Tso and her team have brought an ancient Tibetan manuscript back to life.
High-definition digital scans preserve every detail of "The Four Treatises of Tibetan Medicine" -- from faded water stains and scholarly marginal notes to ancient Tibetan characters -- in this handwritten Ming Dynasty (1368-1644) copy of the renowned medical classics.
With precise AI algorithms, Yong Tso, a professor at the university's school of information science and technology, analyzes and digitizes aging documents page by page, giving sustainable and widespread access to ancient Tibetan medical wisdom.
This year, Yong Tso received the title of National May 1 Labor Medal, one of China's highest honors for workers.
Digital Tibetan language services have become ubiquitous in Xizang. Elderly residents book medical appointments via Tibetan healthcare systems, monks send Tibetan text messages on mobile phones, and herdsmen use voice-to-text input to chat with family on WeChat.
These seamless, everyday conveniences feel ordinary today, yet few notice that the digitalization of the 1,300-year-old Tibetan language only began to take shape over the past three decades, driven by the efforts of the government, as well as researchers like Yong Tso.
Growing up in the regional capital Lhasa, Yong Tso was admitted to study computer science and technology at Shaanxi Normal University in northwest China's Shaanxi Province in 1994, when computers were still a rarity.
"Only in computer class could I use one, and booting it up required typing a long, complicated command," she recalled.
Once, as she stared at the blinking cursor on the screen, it occurred to her: "The English operating system works smoothly. What if Tibetan could also 'run' on a computer? Wouldn't the ancient scriptures in monasteries and libraries be able to 'come alive' on screen as well?"
The vision began to take shape in 1998, when Yong Tso joined Xizang University as one of the earliest faculty members in the school of information science and technology. She became part of the research team led by Nyima Tashi, a leading figure in Tibetan information technology and an academician of the Chinese Academy of Engineering.
The team had released the official international encoding standard for Tibetan a year earlier, laying the groundwork for large-scale digitization of ancient Tibetan texts. Even so, enormous challenges remained, including mainstream systems' incompatibility with ancient Tibetan variant characters and special symbols that blocked ancient document archiving, primitive hardware and crash-prone systems that slowed down research.
Yong Tso immersed herself in the technical work, tirelessly refining Tibetan recognition technology across different scenarios. Step by step, they charted the technical path to make Tibetan "seen" and "recognized" in the digital world.
As the tide of history surged forward and internet and AI technologies evolved at breakneck speed, Yong Tso left the plateau several times for further study. In 2003, she went to Norway for advanced training and later participated in research projects and academic exchanges in the United States and Canada.
It was in 2017 that she truly got involved in digitizing ancient Tibetan manuscripts, when she led the university's first national key research and development project.
The biggest challenge was the interdisciplinary divide. "Those who understood algorithms didn't know the ancient texts, while those who knew the texts didn't understand algorithms," she said.
Instructed by Nyima Tashi, the team, which includes experts in computer science, Tibetan language and literature, and Tibetan medicine, developed a full-process workflow for digitizing ancient Tibetan texts, tackling interdisciplinary challenges through repeated practice.
They successfully developed a Tibetan document layout analysis and multi-font text recognition system, which achieved over 95 percent accuracy on standard printed texts and over 86 percent on handwritten texts.
The technology has been applied to the national ancient text preservation project at the Potala Palace, enabling endangered scriptures to be permanently preserved in digital form while supporting medical knowledge extraction and advancing Tibetan medicine research and cultural interpretation.
For 27 years, Yong Tso has mentored over 40 post-graduate students pursuing a master's or doctoral degree. Over 1,000 Tibetan-Chinese bilingual information technology professionals have graduated from the university.
The team led by Nyima Tashi has achieved a series of systematic breakthroughs, from the Tibetan version of the Kylin operating system to national key R&D programs and major national AI projects in areas such as Tibetan language intelligence, automatic recognition, ancient text analysis, Tibetan medicine digitalization, and a Tibetan large language model.
The Chinese government has been supporting Tibetan language informatization. At the end of 2015, the national standard "Information Technology -- Vocabulary" in Tibetan was officially released, becoming the country's first national standard vocabulary for information technology in an ethnic-minority language. In 2023, an online platform to deal with queries concerning Tibetan and Mandarin was launched, hosting a database of 300,000 standard terms, according to a white paper released last year.
"With modern information technology greatly improving text recognition accuracy, the medical wisdom encoded in 'The Four Treatises of Tibetan Medicine' has been digitally revived and made accessible once more," Yong Tso said. ■
