AudioAid

Inspiration As biomedical engineering students, we knew that hearing loss affects over 1.5 billion people worldwide, and one of the most common complaints from hearing aid users is that their devices make music sound worse, not better. Hearing aids are optimized for speech, not music. We wanted to build something that specifically addresses that gap. What it does AudioAid lets users take a clinical-style hearing test directly in the browser. It plays pure tones at 6 standard audiometric frequencies and records the volume at which the user can first hear each one. From that, it generates a personal hearing profile and applies a real-time frequency compensation to any uploaded song, boosting exactly what the user's ears are missing so music sounds fuller and clearer. How we built it We built a React and TypeScript frontend with Tailwind CSS for styling. The audio engine runs entirely in the browser using the Web Audio API, with a chain of peaking EQ filters at the standard audiometric frequencies (250Hz to 8kHz). The hearing test uses an oscillator node with a gradually ramping gain to simulate clinical pure tone audiometry. The backend is a Node.js and Express server with in-memory profile storage. Challenges we ran into Getting the Web Audio API filter chain to apply smoothly without audio clipping or popping took careful tuning. We also had to handle the async nature of the hearing test state carefully to ensure thresholds were recorded correctly across all 6 frequencies. Accomplishments that we're proud of We built a real in-browser audiometric hearing test from scratch using pure Web Audio API. The compensation engine is clinically grounded, not just a generic EQ. As BME students presenting at an audio hackathon, we're proud thatour biomedical background gave us a depth of clinical knowledge that made this project genuinely different. What we learned We learned how the Web Audio API works at a deep level, how real audiograms translate into EQ compensation curves, and why hearing aids genuinely struggle with music compared to speech. We also learned a lot about building and debugging a full stack app under time pressure. What's next for AudioAid We want to add real audiogram file import so users can upload their clinical audiogram directly from their audiologist. We also want to explore integration with music streaming services and build a mobile version so people can use it with any headphones on the go.

Built With

• and
• api
• audio
• core
• css
• engine
• for
• node.js-and-express-on-the-backend
• powering
• tailwind
• the
• we-built-audioaid-using-react-and-typescript-on-the-frontend
• web
• with