Sunday, December 08, 2024

From Istanbul’s libraries to digital realm: Reviving Ottoman legacy with AI

How a student-led initiative is using artificial intelligence to preserve and digitise millions of Ottoman-era texts, bringing a forgotten literary heritage into the modern age.


Edibe Beyza Caglar
TRT/AA


AA

Vesiqari has created a digital platform that makes Ottoman Turkish texts available in modern Turkish using the Latin alphabet. / Photo: AA

From the heart of Istanbul, nestled among its storied bookshops and cobblestone streets, a group of university students embarked on a mission to breathe new life into the Ottoman language.

Their project, “Vesiqari,” uses artificial intelligence to bridge centuries of history, making the Ottoman Empire’s literary treasures accessible to modern audiences.

What began as a casual stroll through the historic bookstores of Istanbul’s Cagaloglu district led to a groundbreaking initiative: the digitisation and translation of millions of Ottoman Turkish texts into modern Turkish using the Latin alphabet. An effort to preserve these invaluable cultural artefacts for future generations.

Ranging from historical chronicles and religious treatises to epic poetry and scientific texts, the Ottoman Turkish texts offer a window into the empire’s rich cultural and intellectual heritage.

Among the enduring classics are Seyahatname by Evliya Çelebi, a vivid travelogue capturing his journeys across the empire, and Leyla and Majnun by Fuzuli, a timeless love poem.

Through the Vesiqari project - its name derived from the Arabic words "vesiqa" (document) and "qari" (reader), meaning "document reader" – countless works are now accessible in modern Turkish and even English translations.


“One day, I was showing my cousin around Cagaloglu,” Abdullah Tarik Omeroglu, the 27-year-old project’s lead, tells TRT World as he reflects on the pivotal moment in April 2018 that sparked the idea.

He adds: “Then I told him, ‘Here, book-selling is dying, let’s visit some publishers.’ We ended up in a publishing house, where I tried to buy Amak-i Hayal by Ahmet Hilmi of Filibe”.



TRT World

For Omeroglu and his team, the journey has been one of innovation and an unwavering passion for reviving the literary heritage of the Ottoman era.


The early 20th-century philosophical text is an important piece of Ottoman literature that blends mysticism and romantic idealism during a period of cultural transformation.

“The publisher offered me a discounted version with a torn page,” he adds.


At another Istanbul publishing house, Büyüyen Ay Publishing, Omeroglu queried why more books didn’t offer student discounts. Here, his query piqued the publisher's interest, and soon, Omeroglu found himself leaving with two sacks of books—and the start of a collaboration.


The publisher also asked a key question: "Do you know Ottoman Turkish?"


When Omeroglu confirmed he did, the partnership expanded to transliterating Ottoman texts into the modern Turkish alphabet.


The birth of Vesiqari

Two years later, while immersed in these translations— Omeroglu’s roommate, a software engineering student, introduced the idea that would lead to the birth of the project: “We could use artificial intelligence to automate the transliteration process.”


Together with their other roommate, an industrial engineering student, the trio, aged 18-22, formed a team. Six months later, they secured a grant from TÜBİTAK, Türkiye’s Scientific and Technological Research Council.


They began developing Vesiqari at Istanbul Technical University’s Technopark.


"It took us three and a half years, but we finally did it,” Omeroglu says.


Achieving success made them very happy, as Omeroglu recalls: “Starting as students together, achieving something for the first time gave us a deep sense of elation.”


Reviving 7 million pages


Following Türkiye’s Alphabet Reform of 1928, which replaced the Arabic script with the Latin alphabet, Ottoman-era texts became largely unreadable to the general public. Vast archives of historical, scientific, and literary works fell into obscurity.

For Omeroglu and his team, the journey has been one of innovation and an unwavering passion for reviving the literary heritage of the Ottoman era.

“Our goal was to bring these neglected pages back into the world of knowledge and scholarship,” Omeroglu explains.


“Through this effort, millions of pages will now be transliterated.”

“Libraries can contribute their collections to the platform, benefiting everyone, while still maintaining control over their own materials,” Omeroglu explains.


The initiative has created a digital platform that makes Ottoman Turkish texts available in modern Turkish using the Latin alphabet.


Today, Vesiqari boasts over 7 million digitised pages, offering free access to institutions through university databases and affordable subscriptions for individual users.

“Our goal isn’t to profit, but to make these texts accessible.”


The AI-powered system not only offers scholars and readers access to rare historical documents but does so in a way that was once thought impossible. One of their first test cases was Halide Edib Adivar’s Seviye Talip, a 1919 novel exploring personal ambition, love, and societal transformation.


Evolving database

“Using this small dataset, we saw immediate success,” Omeroglu recalls.

“That’s when we realised we could do so much more with a larger database.”

This success proved that their idea had real potential, addressing a long-standing problem: the unreadability and inaccessibility of century-old texts. “Now, these works can be enjoyed by a much larger audience,” he adds.


The beauty of Vesiqari lies in its collaborative nature.


“Libraries can contribute their collections to the platform, benefiting everyone, while still maintaining control over their own materials,” Omeroglu explains.

“If a library prefers not to share its works, it’s not forced to. We’ve found a way to respect privacy while creating a shared space for those who want to contribute.”


One of the most exciting features of Vesiqari is its speed and efficiency.


“I think what makes Vesiqari unique is that the existing database is stored in memory, constantly evolving,” Omeroglu says. “This means that all Ottoman printed documents are brought together, allowing users to perform a single search and obtain the desired text in translation within just one minute.”


The platform’s open-source nature also encourages other institutions to contribute their own collections, further expanding the database and enhancing its capabilities.


Academic praise Dr Hasan Umut, Assistant Professor at Boğaziçi University and an advisor to the project, applauds Vesiqari as a milestone in digital humanities.

Speaking to TRT World, he says: “Vesiqari is an important achievement in integrating Ottoman studies with digital skills. I believe it will be helpful to researchers and those interested in the field. I'm pleased that the project has achieved such success.”


Dr Umut, who specialises in Ottoman intellectual history, sees the platform as a transformative resource. “

Digital humanities in Ottoman studies have seen remarkable progress,” he remarks. For Omeroglu and his team, the journey from casual bookstore visits to technological innovation exemplifies the power of curiosity and collaboration.


Today, Vesiqari stands as a testament to their vision: a digital bridge connecting past and present, preserving the Ottoman Empire’s literary legacy for generations to come.

No comments: