Three 15-year-old students from Santa Clara, California—Akhil Nagori, Evann Sun, and Lucas Shengwen Yen—have developed AI-powered wearable glasses that translate text to speech in real time for under $100, a stark contrast to the billions Silicon Valley tech companies are investing in similar technology. The innovative prototype won the trio a $10,000 award at the prestigious Thermo Fisher Scientific Junior Innovators Challenge in October after five months of development.
The glasses are designed specifically to help visually impaired students by capturing images of text, extracting the content, and converting it to audio through tiny built-in speakers. The device achieves over 90% accuracy with an average load time of just 13 seconds. The inspiration came from Nagori’s visit to India, where he witnessed his visually impaired great-uncle, who works as a cashier, laboriously reading through boxes of braille receipts line by line.
To create their prototype, the teenagers tackled three major challenges: hardware design, software development, and data collection. They custom-designed the glasses frames using Fusion 360 CAD software and 3D printing, incorporating a camera, battery, speakers, and a Raspberry Pi computer board. The design was optimized for middle and high school students to wear comfortably throughout an entire school day.
For the AI software component, the team custom-trained a convolutional recurrent neural network (CRNN) using a dataset of 800 images collected from school textbooks and educational materials. They ensured the model could handle various fonts and colorful images typical of textbooks, testing it under three different lighting conditions: classroom, low lighting, and outdoor environments.
The journey wasn’t without obstacles. The team pulled multiple all-nighters, and hours before their competition presentation, the soldering came off the Raspberry Pi during their flight. In a dramatic last-minute fix, they worked through the night with a soldering iron purchased from a nearby mechanic store to repair the glasses.
Beyond their team award, Nagori received the Thermo Fisher Scientific Leadership Award, while Sun earned $10,000 for the Lemelson Foundation Award for Invention. The boys have now received a $5,000 grant to scale production and are currently manufacturing 30 units in Nagori’s garage to distribute throughout California, demonstrating their commitment to making assistive technology accessible and affordable.
Key Quotes
Our main goal was to create an easy, cost-efficient way to transcribe text from any format for visually impaired students.
Akhil Nagori explained the core mission behind their AI-powered glasses project, emphasizing the focus on accessibility and affordability that distinguishes their approach from expensive commercial alternatives.
He has all these boxes filled with these braille receipts. He has to go through them line by line. When I saw that, I said, ‘There’s got to be an easier way that’s not so tedious.’
Nagori described the personal inspiration for the project after witnessing his visually impaired great-uncle’s struggles as a cashier in India, demonstrating how real-world observations can drive meaningful AI innovation.
One of the most important aspects of our project is the load time. And that averaged around 13 seconds.
Lucas Shengwen Yen highlighted the technical achievement of their AI system’s processing speed, which is crucial for practical, real-time use in educational settings where students need quick access to information.
I think it really taught us that even if we don’t get what we want the first time, as long as we work hard and stay committed, we can come back and be better.
Evann Sun reflected on their journey after initially not receiving a nomination but ultimately winning national recognition, offering an important lesson about perseverance in innovation and research.
Our Take
This story perfectly illustrates the democratization of AI technology and challenges the notion that breakthrough innovation requires massive corporate budgets. These teenagers leveraged accessible tools—3D printers, Raspberry Pi, open-source AI frameworks—to create a solution that addresses a genuine social need. Their success raises important questions about the AI industry’s resource allocation and whether billion-dollar investments are always necessary for meaningful progress.
What’s particularly impressive is their end-to-end approach: custom hardware design, neural network training with domain-specific data, and real-world testing under various conditions. This demonstrates sophisticated understanding of the complete AI product development cycle. Their focus on battery life, accuracy, and load time shows mature product thinking often missing in academic projects.
The $5,000 grant for scaling to 30 units represents the next critical challenge: moving from prototype to production. If successful, this could establish a blueprint for affordable assistive AI technology and inspire similar youth-led innovations addressing accessibility gaps worldwide.
Why This Matters
This story represents a significant moment in democratizing AI technology and highlights how young innovators can create practical, affordable solutions that rival corporate investments worth billions. While major tech companies pour massive resources into developing wearable AI products, these teenagers proved that accessible AI innovation doesn’t require enormous budgets—just creativity, determination, and technical skill.
The project addresses a critical accessibility gap in education and daily life for visually impaired individuals. Current assistive technologies are often expensive and inaccessible to many who need them most. By creating a sub-$100 solution with 90% accuracy, these students have demonstrated that AI-powered assistive technology can be both effective and affordable.
This achievement also showcases the growing sophistication of AI education at the youth level and the potential for the next generation to drive meaningful innovation in artificial intelligence. Their use of custom-trained neural networks, edge computing with Raspberry Pi, and real-time text-to-speech conversion demonstrates advanced technical capabilities. The story serves as inspiration for young STEM students and highlights how AI can be leveraged for social good and inclusive design, potentially influencing how the broader tech industry approaches accessibility challenges.
Related Stories
- Sam Altman and Jony Ive Partner on AI Device, Target $1B Funding
- Harvard Students Behind Facial Recognition AI Eye Construction Industry
- Artificial Intelligence (AI) in Healthcare Market Outlook 2022 to 2028: Emerging Trends, Growth Opportunities, Revenue Analysis, Key Drivers and Restraints
- Big Tech’s 2025 AI Plans: Meta, Apple, Tesla, Google Unveil Roadmap