Open-source AI models can help bridge linguistic barriers in India, enabling academic learning in local languages and fostering linguistic inclusivity. Governments can leverage these AI models to build personalized tutoring in Indian languages.
Open-source foundational models can help with automatic voice-to-text translation, document translation, and knowledge dissemination in India. These models can bridge linguistic barriers without compromising cultural identities, enabling academic learning in local languages and nurturing linguistic inclusivity. The government plays a pivotal role in investing in policies, incentives, content creation, and capacity building for the enablement and adoption of Indian languages. This includes providing learning resources, teacher training, technology solutions, foundational models in Indian languages, research promotion, and raising awareness.
Companies like DheeYantra and Saarthi.ai are making strides in the Indian language AI/NLP conversational platform and automation, while Devnagri.ai, Karya.in, and reverie offer voice localisation and human translation platforms in vernacular languages. Governments can leverage their services to translate vast amounts of content, enriching the pool of data available in Indian languages.
Seed investment can also be made in leveraging these foundational models to build personalized tutoring in Indian languages. Khan Academy’s Khanamigo chatbot is an example of this. Infrastructure through skill hubs and common service centers can mitigate resource scarcity. The Commission of Scientific and Technical Terminology (CSTT) is developing dictionaries for science and technology using Sanskrit roots and grammar, leveraging ancient Indian knowledge for cutting-edge technologies.
A policy framework is needed to adopt a lifecycle of teaching and learning in Indian languages. The National Education Policy will see widespread adoption only once enabled through the National Credit Framework. Opportunities in the market are a significant pull factor towards the use of Indian languages, with AI-driven solutions such as Fasal Kranti providing real-time data to farmers in Maharashtra in Marathi or recruitment platforms like Dhiyo.ai offering voice-based hiring services using AI/ML techniques. Innovations like Ujjivan Small Finance Bank’s Hello Ujjivan provide India’s first mobile banking application with 3 V’s – voice, visual, and vernacular-enabled features – to provide banking access to individuals with limited reading and writing skills.
Conclusion
Open-source AI models can help with automatic voice-to-text translation, document translation, and spreading knowledge in India. These AI models can bridge linguistic barriers and enable academic learning in local languages, fostering linguistic inclusivity. The government plays a crucial role in investing in policies, incentives, content creation, and capacity building for the enablement and adoption of Indian languages. Companies like DheeYantra and Saarthi.ai are making strides in the Indian language AI/NLP conversational platform and automation, while Devnagri.ai, Karya.in, and reverie offer voice localisation and human translation platforms in vernacular languages. Governments can leverage their services to translate vast amounts of content, enriching the pool of data available in Indian languages.
Personalized tutoring in Indian languages can also be leveraged, with Khan Academy’s Khanamigo chatbot and the Commission of Scientific and Technical Terminology (CSTT) developing dictionaries using Sanskrit roots and grammar. A policy framework is needed to adopt a lifecycle of teaching and learning in Indian languages, with operational guidelines and mechanisms being created. Market opportunities, such as AI-driven solutions and mobile banking applications, also play a significant role in promoting the use of Indian languages.
