Skip to content
Projects

lgordey/typing-reply has 10 GitHub stars and 0 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-18 17:57 UTC. Last push 2026-11-16.

How these counts are measured

Measured star growth

in 24h
0
in 7 days
Collecting history
in 15 days
Collecting history
in 1 month
Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

About

Reply to voice messages with a video of hands typing your answer. Rendered on-device with WebCodecs.

Creative & Media

README.md

Original

please just text me

pleasejusttext.me — got another voice message? Reply just as inconveniently: type your answer and get a video of hands slowly typing it on a phone. To read it, they’ll have to watch.

Available in English, Russian, Spanish and French.

How it works

Everything happens in the browser, on the user’s device — there is no backend that sees the text.

  1. Timeline — the text is turned into a deterministic sequence of key presses on an iOS-like keyboard (shift, caps lock, 123 page, 🌐 layout switch, long-press accents, typos with backspace, grandpa’s one-finger hunting). Seeded PRNG, so the same seed gives the same video.
  2. Renderer — renderFrame(ctx, timeline, t) draws one frame into any Canvas 2D context. It has no DOM dependencies, so the same code runs in Node via @napi-rs/canvas.
  3. Encoder — frames go one by one into the browser’s H.264 encoder (WebCodecs) and are muxed into MP4 with Mediabunny. Nothing is kept in memory, which matters on iOS Safari.
  4. Share — the Web Share API hands the file to WhatsApp, Telegram or any messenger.

The only network calls are anonymous usage counters (POST /api/event → Cloudflare Workers Analytics Engine: event name, mode, language, text length, render time) and Cloudflare Web Analytics. No cookies.

Development

npm install
npm run dev          # http://localhost:5173
npm run dev:phone    # HTTPS on your LAN — WebCodecs and Web Share need a secure context
npm test             # timeline/keyboard tests + renderer smoke test in Node
npm run frames -- "Hello there" en normal "1,3,end"   # render frames to out/frames/*.png
npm run og           # regenerate public/og.jpg and the touch icon

Requirements for the generated video: iOS 16.4+ Safari or a recent Chrome (WebCodecs with H.264).

Project layout

Path What
src/timeline/ text → key-press timeline (modes, typos, limits)
src/keyboard/ iOS-like layouts (en, ru, es, fr) and the keyboard simulator
src/render/ frame renderer: phone, chat, keyboard, hands, end card
src/encode/ WebCodecs + Mediabunny MP4 encoder
src/i18n/ UI and in-video strings
worker/ Cloudflare Worker that receives analytics events
scripts/ frame preview, OG image, npm run stats funnel report

Deploy

Static site + a tiny Worker on Cloudflare (wrangler.jsonc). Pushes to main deploy the dev environment; production is deployed manually via the Deploy production GitHub Action.

Contributing

Translation fixes and ideas are very welcome — open an issue or a PR, or write to [email protected].

License

MIT