EchoTake macOS · Coming soon

TL;DR — the tool that finally gets those screencasts made

Screen recording
that knows what happened.

Recording a demo of your own app is painful enough that it rarely happens: takes have to be perfect, updates make them stale, every language is a new production. EchoTake records the screen plus every click, scroll and keystroke as data — cut freely, re-record against the live app, export every language from one edit.

In development macOS 14+ Media Atelier
Take 01 · cursor track

01 — The cursor track

The pointer is never filmed.

EchoTake captures the cursor as a separate data track and draws it synthetically when the film is rendered.

  • After a cut, the cursor is interpolated from the previous click to the next — takes don't have to be perfect.
  • Raw paths during drags, no jump at cuts.
  • Pointer size, style and motion are render settings — decide them after the recording, not before.
PIXEL RECORDING jump at the cut ECHOTAKE interpolated · click to click

The same cut, filmed vs. drawn.

02 — Editing & camera

Nondestructive editing.

Recordings are never modified. Cuts, zooms and retiming are render instructions — reversible weeks later.

  • The zoom camera frames clicks and interactions, not pointer motion.
  • Automatic targets, adjustable per keyframe.
  • Optical-flow slow motion.
FRAMED ON THE CLICK
Re-Record

One edit. Every language.

EchoTake replays the recorded session in your app — AX anchors, resolved through its string tables, time-locked to the original take. Switch the app to French, hit Re-Record: the same film in French, edit intact.

Step 01

Record the session once

Clicks, menus, scrolling and typing are stored as events alongside the video.

Step 02

Cut it once

Trim, zoom, title, score — one edit for all language versions.

Step 03

Re-Record in any language

Anchors resolve against the localized UI; the app performs the take again; the edit stays.

ENDEFRESITNLJAPT+

03 — Narration

Narration, per block.

Narration is written and voiced per text block, so you fix one sentence without re-recording the whole track.

  • Synthetic voices from Google, Apple's on-device engine, or plugins such as ElevenLabs.
  • Or record your own voice, block by block, with the built-in teleprompter.
  • Translation runs against the UI glossary of the app on screen — Claude or Apple's on-device models — so the narration says exactly what the interface says. No invented menu names.
BLOCK 01 GLOSSARY MATCH “Export Selection…” „Auswahl sichern…“ the words on screen · the words you hear

Translation uses your string tables.

04 — Finish

Titles, sound, slow motion, export.

01

Titles

Template-based title cards: font, animation, crossfade.

02

Click effects

Rendered from the click track.

03

UI sounds

Placed automatically on the recorded events.

04

Ambient music

Looped beds, ducked under narration.

05

Slow motion

Optical-flow retiming, down to 0.2×.

06

Export

One project, every language version, rendered from the same edit.

05 — Extensible

A plugin is one file.

Voice and translation services are single JavaScript files with an .echoservice extension. Double-click to install. No bundles, no installers, no restart.

// ElevenLabs.echoservice
service = {
  name: "ElevenLabs",
  capabilities: ["tts"],
  settings: [{ key: "api_key", secure: true }],
  async synthesize({ text, voice }, s) { /* http(…) */ }
};

06 — MCP-first

Built to be driven by an agent.

Every operation is an MCP tool: record_start, timeline_split, clip_rerecord, export_clip and the rest. An agent can produce a finished screencast end to end.

  • Batch a release's demo videos from one instruction.

Coming soon

EchoTake is in development.

For macOS 14 and later. Write to us and we'll tell you when the first build is ready.

info@mediaatelier.com

No newsletter · No tracking · Just an email