SnapVoice: screenshot-to-video guides

Turn scattered screenshots plus a voice note into a vertical how-to video for your parents

Problem

Explaining small household tasks to parents over a video call is tedious: by the time you finish describing where each decoration goes, they have forgotten. Recording separate voice notes per screenshot is also too much work, and long text or spreadsheets never get read.

Solution

You record one voice explanation over a set of screenshots, and on-device AI transcribes it into subtitles, then aligns each subtitle with the right image and exports a vertical short video. Everything stays on the local machine, and the result is ready to send over WeChat.

Users

Adults explaining daily life tasks to parents remotely; small team leads making how-to notes; craft and tutorial creators

MVP

Import and reorder screenshots; record voiceover and transcribe locally with Whisper; auto-split images by subtitle timing; export a 9:16 video with templates; share directly to messaging apps

Revenue

Free exports with watermark; one-time unlock for watermark removal and HD export; creator plan with export quotas

Why now

On-device speech recognition (Whisper, Gemini Nano) now runs offline on ordinary laptops and phones, so this lightweight tool no longer needs backend servers and can be prototyped in a weekend