Cassette

Homepage
Application

Inspiration

Cassette was inspired by conversations with Marcus, the blind receptionist at the California School for the Blind. We often talked about movies, and it sparked a question I couldn't stop thinking about: What if blind people could direct films themselves? That idea stuck with me, and later I discovered Audiojack, a sound-only storytelling platform from the American Printing House. As a film enthusiast and a longtime advocate for assistive technology, I saw an opportunity to merge both passions, and Cassette was born.

What it does

Cassette is a voice-controlled audio film editor designed specifically for blind and visually impaired creators. Users can build immersive sound-based stories by combining music, sound effects, and dialogue through natural speech. Cassette uses text-to-speech to generate dialogue with customizable tone and accent, integrates with the Freesound API for audio assets, and allows timeline control through voice commands alone — no screens or visuals required.

How I built it

Cassette is built as a web application using HTML, CSS, and JavaScript. I integrated the Freesound API for sound effects and music, and used the Web Speech API for both speech recognition and text-to-speech playback. Dialogue and audio events are structured as timeline components that can be played, moved, or layered using simple spoken instructions. I focused heavily on making the experience fully non-visual, testing each interaction from an accessibility-first perspective.

Challenges I ran into

Designing a fully voice-driven interface was a major challenge, especially ensuring smooth, intuitive interaction for timeline editing without visual feedback. Managing timing, overlapping tracks, and playback with just voice required careful UX design and testing. I also faced limitations with available text-to-speech options in terms of emotional tone and voice variety, which I had to work around creatively.

Accomplishments that I'm proud of

What I learned

Accessibility isn't just a feature — it’s a creative framework. I also deepened my understanding of voice interaction design, browser-based speech APIs, and how to break down complex creative workflows into intuitive, spoken commands.