29/09/2023
India's higher education system has long been characterized by the dominance of the English language as the medium of instruction, leading to a significant barrier for students whose primary language is not English. Recognizing the need to promote education in local languages and make it more accessible to a diverse population, the University Grants Commission (UGC) of India introduced guidelines for translating higher education textbooks using AI tools like ChatGPT.
This article delves into the role of AI and machine translation tools in supporting India's indigenous language policy, their impact on students and institutions, ongoing projects, challenges, and the evolving role of AI in reshaping the landscape of higher education.
The predicament described above has since sparked a discourse among countless Indian students whose mother tongue is not English. The language barrier is a roadblock to their understanding of complex subjects and their grasp of intricate concepts. The burden of navigating this linguistic chasm leads to exclusion and alienation among Indian students from non-English-speaking backgrounds. This issue goes beyond the classroom as it permeates the broader societal and economic fabric of India. The ability to communicate effectively in English is frequently viewed as a ticket to better job prospects and social mobility, intensifying the pressure on students.
UGC has taken steps to resolve these issues by allowing exams in local languages. It has become a crucial step toward linguistic inclusivity under the push of India's language policy to integrate the mother tongue into academic institutions. Besides this, the UGC has now implemented the need to translate English-language textbooks and study materials into many Indian languages within higher education institutions. It is a monumental task, which is why it is no wonder that the policymakers and officials in UGC have sought the use of Artificial Intelligence (AI) and machine translation to shape the landscape of their multilingual educational system.
This year, the UGC issued comprehensive guidelines for AI-assisted textbook translation. These guidelines laid out the framework for translating educational materials from English into various Indian languages using AI. They emphasized the importance of maintaining consistency and quality in translated textbooks, ensuring students receive accurate and accessible content in their mother tongue.
AI translation tools, such as ChatGPT, have the potential to bridge these linguistic gaps by enabling the translation of educational content into numerous languages. India is no stranger to Artificial Intelligence as it has an expansive landscape of Indian technical education. The All India Council for Technical Education (AICTE) has emerged as a trailblazer, spearheading innovative initiatives to redefine how students access and engage with educational content. Under the AICTE, they have begun to adopt AI translation tools significantly. It has undertaken projects to translate Bachelor of Technology (B.Tech) textbooks into local languages, making technical education more accessible to students across India.
One notable initiative is 'Anuvadini,' an AI-powered translation tool developed by AICTE. Unlike other AI systems, Anuvadini supports 22 regional Indian and foreign languages. It was built to break language barriers and unify India and can be used to create voice and document content. Based on the UGC's guidelines, the use of Anuvadini and other AI tools can be used in the "first round of translation," but it must be followed by thorough manual post-editing to ensure the accuracy of the translated content. The level of the tool's accuracy differs for each subject and language. And so if the machine translation and AI tools translate the text word-for-word, it must be rephrased to create meaningful content.
Besides Anuvadini, several top engineering colleges in India have embarked on research projects in generative AI to support translating educational materials. For instance, Project Udaan at IIT Bombay focuses on translating scientific and technical content into various regional languages, making STEM education more accessible.
AI translation tools represent a remarkable leap forward in breaking down language barriers and fostering global communication. Yet, their potential is not without its unique set of challenges, and addressing them is pivotal to harnessing the true power of AI in the realm of translation.
1. The Role of Authentication: While AI translation tools exhibit impressive capabilities, they are not infallible. Human intervention, in the form of trained professionals, is indispensable for authenticating translations to ensure accuracy and quality.
2. The Human Touch: Language is not just about words. It's about context, cultural nuances, and the intricacies that machines might overlook. Trained professionals bring their linguistic expertise and cultural understanding to the table, ensuring that the translated content maintains linguistic fidelity and cultural sensitivity, adding a layer of depth and accuracy that machines alone cannot achieve.
3. Technical Terms and Specialized Content: Technical terms, domain-specific jargon, and intricate scientific concepts can defy straightforward translation. Machine translations may require additional help to accurately convey these specialized nuances, underscoring the need for manual intervention.
4. Pan-Indian Common Core Vocabulary: It is essential to encourage the development of a pan-Indian common core vocabulary to promote effective multilingual education. This approach transcends mere word-to-word translations, aiming to convey complete concepts. This common core vocabulary becomes the linchpin of effective multilingual communication by standardizing terminology and promoting a shared understanding across languages.
5. Biases in Machine Translation and AI Systems: Many research have shown that machine translation and AI programs run the risk of obtaining biased information and data that can harmfully affect marginal communities and perpetuate detrimental ideas about certain groups of individuals. Language professionals and AI experts must oversee the gathering and collecting of data to ensure that these harmful biases will not cause issues with India's plans of inclusivity and cultural and linguistic diversity in the academe.
Much like India, Abu Dhabi has also begun to implement and create an AI tool for its Arabic communities. These AI initiatives just show how transformative and revolutionary this technology has become. Low-resource languages now have the opportunity to make their language more accessible to the public, particularly in India's educational system.
These tools have the potential to break down language barriers, making education available to a more diverse student population. However, their successful implementation hinges on continued development, quality assurance, and the collaboration of AI and human educators. As India continues to prioritize indigenous languages in education, AI translation tools will play an increasingly vital role in reshaping the future of higher education in the country.