AI Editor for YouTube Shorts

I need an AI-powered workflow that turns raw vertical footage into polished YouTube Shorts in one pass. The core requirement is reliable automatic captions and subtitles that are timed accurately, styled to match brand colours, and export cleanly in 9:16, ≤60-second MP4 files.

Here’s how I picture the process: I drop a clip (up to 2 GB, 1080×1920), the system detects speech, generates the transcript, burns in the captions or exports an .srt if I choose, then delivers a ready-to-post Short with minimal manual correction. If you already have a proven model or can stitch together Whisper / ffmpeg / moviepy / Premiere Pro scripting—or you prefer to build a lightweight React or Electron front-end that calls cloud functions—I’m flexible on the tech stack as long as the output quality is TikTok/Instagram-level crisp.

Acceptance criteria
• Captions sync within ±300 ms of spoken words
• Custom font, size, colour and safe-area options
• One-click export to 1080×1920 MP4, ≤60 s,