Amidst the digital revolution in India, innovative AI tools are enhancing communication for delivery drivers and beyond, ensuring inclusivity for speakers of the nation’s 22 official languages.
Unlocking AI Accessibility in 22 Languages: A Step Toward Inclusion
Unlocking AI Accessibility in 22 Languages: A Step Toward Inclusion
New AI translation initiatives aim to bridge language barriers in India's diverse workforce.
In the bustling streets of Mumbai, delivery driver Vineet Sawant shares his experience navigating work challenges in a multilingual environment. As a native Marathi speaker with limited English proficiency, Mr. Sawant initially struggled with his job at Zepto, an online grocery delivery service. However, life on the road has become significantly easier since the company partnered with Reverie Language Technologies last year, launching an AI-powered translation service. Now, thanks to the Zepto app, drivers can select their preferred language, eliminating the confusion of reading delivery instructions.
"At first, it was difficult," says Mr. Sawant, reflecting on his early days. "Everything was in English, and I could understand some of it, but I'm far more comfortable in Marathi." This innovation has not only streamlined communication but has also enhanced productivity, allowing Mr. Sawant to triple his daily deliveries.
India's rich tapestry of languages presents both opportunities and challenges. Professor Pushpak Bhattacharyya from IIT Mumbai highlights the diversity of the nation, pointing out that without accessible technology, millions risk being excluded from vital sectors including education, healthcare, and governance. The rollout of systems like ChatGPT has made it clear that AI language inclusivity remains urgent as the demand for robust data grows.
While widely spoken languages like Hindi and English can easily find extensive data, many lesser-known languages and regional dialects lag due to data scarcity. Professors and tech pioneers are actively seeking solutions to this issue, with projects like Bhashini—an initiative aimed at creating quality datasets and AI language models in 22 languages. Launched in 2022, it has made considerable strides, working with multiple government departments to facilitate public services through multilingual chatbots and increased language accessibility.
Amitabh Nag, CEO of Digital India’s Bhashini Division, emphasizes the project's significance, stating, "Bhashini ensures India's linguistic and cultural representation by building India-specific AI models rather than relying on global platforms." The goal is to empower rural users with voice-enabled access to essential services in their native tongues within a few years.
Research is ongoing to expand AI’s reach beyond translation, including an innovative project by Kshitij Jadhav from the Koita Centre for Digital Health. Jadhav is developing an AI model to assist individuals in quitting smoking, tailoring advice based on their unique needs and language, a task made challenging by the lack of multilingual healthcare professionals. Initial developments are focusing on English and Hindi, with aspirations of offering support across all 22 languages.
Back in Mumbai, Vineet Sawant’s storytelling is a testament to the profound effect of these technological advancements. “When the app speaks our language, we feel more confident, and we work better,” he reflects. As AI continues to evolve, its commitment to serve all linguistic communities in India could pave the way for a more inclusive future.
"At first, it was difficult," says Mr. Sawant, reflecting on his early days. "Everything was in English, and I could understand some of it, but I'm far more comfortable in Marathi." This innovation has not only streamlined communication but has also enhanced productivity, allowing Mr. Sawant to triple his daily deliveries.
India's rich tapestry of languages presents both opportunities and challenges. Professor Pushpak Bhattacharyya from IIT Mumbai highlights the diversity of the nation, pointing out that without accessible technology, millions risk being excluded from vital sectors including education, healthcare, and governance. The rollout of systems like ChatGPT has made it clear that AI language inclusivity remains urgent as the demand for robust data grows.
While widely spoken languages like Hindi and English can easily find extensive data, many lesser-known languages and regional dialects lag due to data scarcity. Professors and tech pioneers are actively seeking solutions to this issue, with projects like Bhashini—an initiative aimed at creating quality datasets and AI language models in 22 languages. Launched in 2022, it has made considerable strides, working with multiple government departments to facilitate public services through multilingual chatbots and increased language accessibility.
Amitabh Nag, CEO of Digital India’s Bhashini Division, emphasizes the project's significance, stating, "Bhashini ensures India's linguistic and cultural representation by building India-specific AI models rather than relying on global platforms." The goal is to empower rural users with voice-enabled access to essential services in their native tongues within a few years.
Research is ongoing to expand AI’s reach beyond translation, including an innovative project by Kshitij Jadhav from the Koita Centre for Digital Health. Jadhav is developing an AI model to assist individuals in quitting smoking, tailoring advice based on their unique needs and language, a task made challenging by the lack of multilingual healthcare professionals. Initial developments are focusing on English and Hindi, with aspirations of offering support across all 22 languages.
Back in Mumbai, Vineet Sawant’s storytelling is a testament to the profound effect of these technological advancements. “When the app speaks our language, we feel more confident, and we work better,” he reflects. As AI continues to evolve, its commitment to serve all linguistic communities in India could pave the way for a more inclusive future.
