The idea came from a simple frustration — I was making music with AI tools and enjoying every part of the process except the last step: getting captions onto the video before posting. It always meant manually syncing timestamps, fixing mishears, and exporting from three different tools. The lyric-paste feature was the core insight. Generic transcription struggles with AI-generated vocals — it mishears words that aren't in its training distribution. But if you already have the lyrics (and with AI music, you usually do), the problem becomes alignment, not transcription. So I focused there first. The goal was to get from audio file to captioned video in under a minute, without touching a timeline editor. I think we're there. Would love to hear what the PH community thinks — especially anyone using Suno, Udio, or similar tools. Your feedback will directly shape what gets built next.