From Speaking Bhojpuri at Home to Working on AI: Why Regional Languages Belong in the Future of Technology
I never imagined that speaking my mother tongue could become a professional skill in the world of artificial intelligence. I'm from Bihar, India. Bhojpuri is a language I grew up hearing and speaking. Like many people w
I never imagined that speaking my mother tongue could become a professional skill in the world of artificial intelligence.
I'm from Bihar, India. Bhojpuri is a language I grew up hearing and speaking. Like many people who grow up with regional languages, there were times when I hesitated to speak it openly in certain settings.
I never thought much about it. It was simply the language of home, family, conversations, humour and everyday life.
Until I got an opportunity to work on Bhojpuri language projects for AI.
My first step into language technology
I started working on Bhojpuri business copywriting, contributing more than 5,000 words to an AI language project.
Later, I got the opportunity to participate in AI evaluation work. This involved evaluating AI-generated conversations for aspects such as naturalness, linguistic quality, cultural appropriateness and helpfulness.
I have now completed 15 evaluation tasks and additional Bhojpuri prompt-editing work.
It has been a fascinating experience to see how language, culture and technology intersect.
The challenge of regional languages in AI
When we talk about AI, we often discuss English, large language models, coding benchmarks and computational capabilities.
But there is another important question:
How well does AI understand the languages people actually speak in their everyday lives?
India has an extraordinary linguistic diversity. Many languages carry unique expressions, cultural references, idioms, humour and ways of communicating.
A model might produce grammatically understandable text while still missing the cultural context or natural expressions of a language.
For languages like Bhojpuri, representation in AI should mean more than simply translating words. It should involve understanding how people naturally communicate.
Why this matters to me
The most meaningful part of this experience isn't the payment or the project title.
It is the realisation that a language I once felt shy speaking has professional value in an emerging technology.
I am still learning, and I am only at the beginning of this journey. But this experience has made me more interested in language technology, AI evaluation and the digital preservation of regional languages.
A question for the AI community
What would it take to make AI genuinely multilingual?
Not just capable of translating a sentence, but capable of understanding the cultural context, expressions and natural communication patterns of languages that receive less technological attention.
I would love to hear from people working in NLP, speech technology, language datasets and AI evaluation.
Perhaps the future of AI should not be measured only by how well it understands the world's most widely represented languages, but also by how thoughtfully it serves the languages spoken closer to home.
Bhojpuri deserves a place in that future.
Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes β full credit and traffic to the original publisher.