Hyderabad: The Ministry of Tribal Affairs has launched an AI-powered translator for tribal languages. Titled Adi Vaani, the platform in its beta version supports Santali, Bhili, Mundari and Gondi languages.
The next phase of the project aims to include more languages like Kui and Garo.
Combined effort by premier institutions
The app has been developed by a consortium of premier institutions led by IIT Delhi and comprising IIIT Hyderabad, BITS Pilani, and IIIT Nava Raipur in collaboration with the Tribal Research Institutes (TRIs) in Jharkhand, Odisha, Madhya Pradesh, Chhattisgarh, and Meghalaya.
According to a press note, the app is available on the Play Store as well as a dedicated web platform.
Can languages go extinct?
As per the 2011 Census, India is home to 461 tribal languages spoken by the Scheduled Tribes and 71 distinct tribal mother tongues.
Among these, 81 are vulnerable and 42 are critically endangered. They face the risk of extinction due to limited documentation and intergenerational transmission gaps.
What does the app do?
Adi Vaani aims to address this challenge by leveraging AI for systematic digitisation, preservation and revitalisation of tribal languages.
The platform does this by enabling real-time translation of both text and speech between Hindi/English and the tribal languages. It also helps in preserving folklore, oral traditions and cultural heritage via optical character recognition technology.
It can support and promote civic inclusion in tribal communities by spreading awareness about governmental schemes and other important initiatives.
Role of IIITH
At IIITH, the Speech and Natural Language Processing groups at the oldest lab of the Language Technologies Research Centre have been involved with the project since July 2024.
“This is a social initiative project that we are proud to be associated with,” said Prof. Radhika Mamidi, who leads the institute’s efforts with the help of Prof. Anil Vuppala and research scholars Vandan Mujadia, Nikhilesh Bhatnagar, Anindita Mondal and Soujanya Rao.
In the initial days, the technology know-how on putting Adi Vaani on the cloud was shared by IIITH’s Product Labs team led by Satish Kathirisetti, along with Sriram and Shashank.
The tech behind the app
The IIITH team used a Transformer-based sequence-to-sequence (seq2seq) architecture for the four machine translation systems of English to Santali, Hindi to Santali and vice versa.
“This has become the state-of-the-art approach in neural machine translation (NMT). The parallel corpus was built with the help of Tribal Research Institute, Odisha. After the base model was built, additional data was generated and post-edited by Santali native speakers, which helped improve the systems,” said Prof. Mamidi.
The researchers also developed a Text-to-Speech (TTS) tool for Santali, Mundari and Bhili languages. The TTS tool for Gondi is currently under development.
Anindita Mondal, who built the TTS tools, worked closely with the native speakers. They spent a considerable amount of time at IIITH for recording speech data.
Promoting inclusive governance
According to the press release, the app is more than just a translation tool; the government aims to safeguard endangered cultures and knowledge.
It seeks to empower the tribal communities with all the public service initiatives in their own languages and promote inclusive governance by ensuring the last-mile reach of governmental schemes.
Dr Radhika Mamidi said, “We will continue to improve the models with more feedback coming in with the Beta launch. Our aspiration is to make NCERT books, educational and health awareness videos, government schemes, and materials to be made available in these low-resource languages using the fast-emerging AI technologies. We also plan to work on more indigenous languages.”
“As part of the Indic-Wiki summer internship programme, under the mentorship of Krupal Kasyap, we focused on enriching online content in Telangana origin languages such as Gondi, Koya, Kolami, Naikdi, Chenchu, Kaikadi (Yerukala), Lambadi, Nakkala and Konda Kammara. Hopefully, we can make AI tools for these languages as well with the support from the Telangana government and the Ministry of Tribal Affairs,” she added.
Janjatiya Gaurav Varsh
The launch was part of ‘Janjatiya Gaurav Varsh’ – a year-long celebration of tribal pride, legacy and empowerment coinciding with the 150th birth anniversary of Dharti Aaba Bhagwan Birsa Munda (a tribal folk hero and independence activist).
