The unfulfilled promise of accessible technology in Bangla

Rafi wakes early in the morning. He reaches for his phone, not to scroll but to listen. Blind since childhood, his ears are his eyes. He taps the screen, and the voice of a text-to-speech (TTS) reader in Bangla begins to narrate the news of the day. Some words come out clear, familiar, comforting. Others arrive twisted, robotic and mispronounced. "Bangladesh" becomes something else entirely. Rafi sighs. Technology is his closest companion, yet it still speaks with a foreign accent in his own language.

Rafi's story is not rare. Across Bangladesh, thousands of people with visual impairments depend on screen readers and speech software. While English-based accessibility tools are more advanced and polished, Bangla TTS and screen-reader systems still struggle with accuracy, fluency and naturalness. Recent developments show that commercial platforms have begun integrating Bangla voices, but they remain inconsistent, particularly on websites or government services. For someone like Rafi, this means every email, job form or e-learning course becomes an unpredictable mix of clarity and confusion.

The same morning, in another part of Dhaka, Maya joins her online class. Maya is not blind; she has low vision. As the lecture begins, her classmates follow along easily, but Maya struggles to enlarge the text on her screen. The built-in magnifier helps only to a point—when she zooms in, she loses the overall layout of the page. She tries to use a screen reader, but Bangla support is inconsistent and often mispronounces key academic terms. A sentence meant to say "the Constitution of Bangladesh guarantees equality" ends up as "Bangladesh constitution guarantee quality." The gap between meaning and understanding widens.

For people with low vision in Bangladesh, technology has opened some doors but left others locked. Tools such as screen magnifiers, text-to-speech software and accessible e-book platforms exist, but they are often poorly localised in Bangla or difficult to configure without proper training. Research institutions have developed prototypes for improved Bangla OCR (optical character recognition) and better magnification systems, but these rarely progress beyond pilot projects. They remain stuck in labs or small-scale trials, while Maya continues to struggle through her classes with tools that are not designed for her needs.

These gaps are not abstract. In a recent multi-year, three-country study I led, exploring the intersectional status, challenges and opportunities of persons with visual disabilities in Bangladesh, we found similar patterns of exclusion. The study—part of the Accessibility, Language and Tech for the People (ALT) project, supported by Whose Knowledge? (WK?), an organisation of anti-colonial feminists working to centre the knowledge of the world's marginalised communities—examined the lived realities of visually impaired people across Bangla, Hindi and Urdu linguistic contexts. ALT highlights how digital technologies and content access remain deeply unequal, especially in local languages, reinforcing the barriers that users like Rafi and Maya face every day.

The stories of Rafi and Maya echo a larger truth: accessible technology in Bangla exists, but it is incomplete, fragmented and often an afterthought.

According to the Bangladesh Bureau of Statistics, about 2.8% of the population—over 4.6 million people—live with disabilities. UNICEF further reports that more than half of children with disabilities do not attend school, largely because classrooms, teaching materials and digital platforms are not designed for them. These figures are not just statistics; they are lived realities, shaping the futures of millions of Rafis and Mayas.

Bangladesh does not lack ambition. Policies are already in place: the Rights and Protection of Persons with Disabilities Act 2013 guarantees equal rights, and government frameworks for digital accessibility exist on paper. But policy without practice is like a locked door with the key missing. Many government websites still fail basic accessibility tests; PDFs remain unreadable for screen readers; mobile apps exclude TTS integration or magnifier-friendly design.

Assistive technology (AT) supply in Bangladesh also suffers from systemic weaknesses. Research shows that while devices such as screen readers, magnifiers and Braille kits are available, the service chain around them is fragile. Delivery ends once the product is handed over. There is little emphasis on user training, customisation or long-term maintenance. For many families, a device becomes an idle object in a drawer rather than a gateway to inclusion.

On the brighter side, Bangla language technology is evolving. Speech-to-text models, Bangla OCR and AI-based TTS are improving. But most remain academic exercises, not mainstream services. The lack of localisation, open APIs and integration with daily-use platforms—schools, hospitals, government portals—means that progress stays siloed. For instance, a high-quality Bangla TTS exists, but government e-services rarely embed it for universal access.

Over the years, organisations such as the Aspire to Innovate (a2i) programme have played a pivotal role in advancing disability inclusion in Bangladesh's digital transformation. Through initiatives such as accessibility audits, the development of localised digital accessibility standards and the promotion of inclusive design in government services, a2i has worked to ensure that persons with disabilities can access essential digital tools—from education platforms to social protection services. It has strengthened digital literacy, trained developers and designers on accessible systems, and fostered partnerships with disability organisations, universities and government agencies. Anchored in global commitments, a2i's work demonstrates how structured, evidence-based interventions can translate policy into practical, nationwide impact, creating a more inclusive tech ecosystem in Bangladesh.

So, what is missing? In my lived experience as a person with visual disabilities and as an expert in disability and inclusion, three major gaps stand out:

The Data Gap: Policymakers do not have robust, up-to-date data on the needs of people with disabilities. Most technology projects rely on assumptions, not user-centred evidence. Without clear demand mapping, investment flows unevenly.

The Service Gap: Assistive technologies are treated as hardware, not ecosystems. Training, repair services and local-language adaptation are often neglected.

The Localisation Gap: Accessible technology in Bangladesh is too often an English transplant. Bangla versions exist but feel like translations, not native solutions. For users, this makes the experience alienating rather than empowering.

What can be done? The way forward requires practical, user-driven action. Here are five urgent steps:

1. Collect better data: National surveys must integrate disability- and technology-related questions. Without understanding how many people like Rafi or Maya need which tools, we cannot design solutions that fit.

2. Strengthen the assistive-tech service chain: Accessibility must be more than delivering devices. Training programmes, local repair centres and user-support networks should be institutionalised, possibly through public–private partnerships.

3. Invest in localisation: Bangla should be at the centre of all accessible technology. Screen readers, TTS, OCR and magnifiers must achieve fluency and accuracy in Bangla, not merely approximate it. This requires open datasets, government incentives and collaborations between tech companies and universities.

4. Expand low-vision support: Tools such as text-to-speech, high-contrast interfaces, screen magnifiers and accessible digital books should be mainstreamed across education, healthcare and government services.

5. Build community-led pilots: Organisations of Persons with Disabilities (OPDs) should be central partners in testing and refining technology. When communities lead, innovations are not only technically sound but socially relevant.

In the end, the story of accessible tech in Bangla is a story of unfinished potential. Rafi's half-correct screen reader and Maya's frustrating magnifier show us the costs of incompleteness. Technology can empower, but only when it is designed with empathy, embedded in policy and localised in the language people live in. For Bangladesh, digital inclusion is not a luxury; it is the next frontier of equity. And to achieve it, accessible technology in Bangla must be more than a side project—it must become the norm. Otherwise, millions will continue to live in a half-open world, where the door to opportunity creaks but never fully swings wide.

Vashkar Bhattacharjee is a person with visual disabilities and a digital accessibility expert and disability-rights advocate with over 20 years of experience

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.