Caption01

Overview

Modern web-based video captioning tool that automatically generates animated captions using AI-powered transcription.

The Problem

Creating animated captions for videos is traditionally a manual, time-consuming process or requires expensive, cloud-based software that compromises privacy. Content creators need a fast, local-first tool to generate professional captions without data leaving their device.

What I Did

Built a fully local browser-based video editor that processes audio using WebAssembly and OpenAI Whisper
Implemented a timeline editor with precise control over caption timing and duration
Created an animation engine using Framer Motion to support presets like bounce, pop, typewriter, and karaoke
Designed a custom dark-mode UI for comfortable long-form editing sessions
Engineered a rendering pipeline to export captions as SRT, VTT, or embedded video

Tech Stack Details

Frontend

Preact, Tailwind CSS, TypeScript

AI Engine

@remotion/whisper-web (WebAssembly)

State Management

Zustand

Animation

Framer Motion

Storage

IndexedDB

Key Learnings

WebAssembly Integration

Learned to optimize AI model performance directly in the browser without server dependency

Video Processing

Gained deep expertise in handling binary media data, timing synchronization, and frame rendering

Performance Optimization

Managed large state objects for timeline interactions to ensure 60fps scrolling and playback