CatchMyWords – Smart Timestamped Search for Audio & Video

🧠 Inspiration

The inspiration for CatchMyWords came from my own experience spending long hours reviewing audio before editing videos. I often found myself scrubbing back and forth just to find that one quote or section I remembered hearing. I wanted a tool that made spoken content searchable, highlightable, and timestamped — just like searching through text.

⚙️ What it does

CatchMyWords allows users to upload audio or video files and search for one or multiple words/phrases within the content. It returns precise timestamps for every match, and users can export those timestamps for use in editing, documentation, or collaboration. The app helps content creators, students, journalists, and researchers instantly navigate long audio files.

🛠️ How we built it

I built this project solo, using:

• Frontend: React, with a focus on intuitive UX
• Backend: Supabase for authentication, user roles, and storage
• Transcription Engine: Assembleyai
• Features:
  • Multi-word search with timestamped results
  • Export functionality (plain text)
  • Keyword highlighting (transcript and playback)

🧱 Challenges we ran into

Working solo meant wearing multiple hats — from design to backend integration. The most difficult parts were:

• Getting audio playback to stay in sync with the transcript
• Making sure search terms were accurately highlighted and tracked across different timestamps
• Linking the timeline to the transcript, allowing for smooth scrolling, zooming, and clickable highlights
• Building search overlap handling for multi-keyword matches
• There are a number of features that we had to be dropped, but overall I believe my MVP serves a purpose

🏆 Accomplishments that we're proud of

• Fully implemented multi-term search with precise timestamp matches
  
• Built a clean export tool that saves users tons of time
  
• Integrated a responsive UI that syncs playback with transcripts and searches
  

📚 What we learned

• Real-world audio processing brings unique UX challenges
  
• The importance of structuring clean data flow between UI, audio, and transcripts
  
• That building tools for your own pain points often leads to something valuable to others
  
• Deepened my knowledge of authentication, audio APIs, and user feedback loops

🔮 What's next for CatchMyWords?

• A Chrome extension that adds timestamped search directly to YouTube and other platforms
  
• Google OAuth and richer signup experience
  
• Better timeline zooming, overlap display, and preview thumbnails
  
• Team sharing features and annotation collaboration
  
• Launching a full pricing system with Starter, Pro, and Enterprise plans

Built With

• assemblyai
• framer-motion
• javascript
• lucide-react
• react
• supabase
• tailwind
• typescript
• vite