Big tech & India work towards language inclusive tech

English language is entering a new sector, prompts for chatbots. Suffice it to say, since prompts are the latest career emerging from ChatGPT like Artificial Intelligence, those who speak English will have a leg up in creating code bases. At least, till ChatGPT becomes proficient in other languages. Attempts to break the language barrier are on and programmers are experimenting with other languages like Russian, Tamil, etc.

In fact, vernacular languages are entering digitization as technology attempts to include more numbers. Microsoft has tied up with CGNet Swara, a citizen journalism portal collaborating with the Gondi-speaking tribal population in central India, to create a Gondi dataset. Also, the tech giant has been trying to add more low-resourced languages, including Idu Mishmi, Mundari, and others through community engagements. This could be a precursor to infusing Indian languages into Microsoft’s platforms like ChatGPT and Bing Chat.

Other big tech companies too are looking to add more languages to their digital portfolios.

In December, Google announced a grant of US$1 million to IIT Madras to open a centre for Responsible AI. The tech giant wants to use language models to bring down bias and enable fair artificial intelligence deployment, particularly in the Indian context.

Read more: Twitter blues: ‘The bird is freed’ but where is it flying to?

Google also tied up with Bengaluru’s ARTPARK and IISc to unveil the AI-based Project Vaani, to compile comprehensive datasets of transcribed text and spoken language from all districts in India intending to make the internet language-inclusive. The project, with the aim of boosting the size and diversity of India’s open-sourced language data, plans to collect over 150,000 hours of curated speech and 100 million sentences of text in Indian scripts.

Google is also working with Indian pharmacists to create an artificial intelligence tool that can translate the illegible scribbles that doctors are notorious for writing, with the help of a phone camera. Users can click a photo of their prescription or upload it from their photo library, which the artificial intelligence can then scan.