Students in Telangana Develop AI Model for Telugu Language

Telangana students are developing a Telugu AI model, contributing data and recordings to preserve the language in AI platforms.

featured-image

Students in Telangana Develop AI Model for Telugu LanguageOn International Mother Language Day, a student-driven initiative in Telangana is gaining momentum to integrate Telugu with artificial intelligence. The project, known as the AI Telugu Yatra, is focused on creating a Telugu-based large language model, similar to ChatGPT. Launched last year as a small-scale effort, the initiative has now expanded to colleges across the state, with students working to bridge the gap between their mother tongue and advanced AI systems.

Participants have contributed over 2,000 hours of recorded audio and documented more than 2.3 crore words and sentences, incorporating regional dialects and cultural narratives. Named “Telugu Satasahasra Yagam,” the initiative is being led by Swecha, a Hyderabad-based organization advocating for open-source technology.



The goal is to develop a Telugu-language AI model that can process and generate content fluently, addressing the lack of regional language representation in mainstream AI platforms. “Creating a dataset for AI in Telugu is a challenge because regional languages are often underrepresented in digital spaces,” Pravin Chandrahas, secretary of Swecha, told Times of India. “To address this, we trained engineering students to collect data using custom-built mobile applications.

Over six months, we conducted AI Yatras in 19 districts, with participation from more than 5,000 students.” As part of the AI Yatra, students are actively involved in promoting the integration of Telugu with AI. They participate in rallies and training programs, contributing both audio and video recordings to document diverse linguistic patterns.

The initiative plans to expand into additional districts in the coming months. “In the first phase, we trained 40,000 engineering students. Now, we are working with degree colleges and aim to train one lakh students,” said Rajashekar of Swecha.

“Students are enthusiastic once they understand the project’s impact.” With the initiative gaining traction, Swecha and student volunteers remain focused on preserving and advancing Telugu in AI, ensuring that the language continues to evolve alongside technological progress..