
Capture Meeting Decisions Without Leaving Your Flow State
Stop losing decisions between back-to-back meetings. A practical capture workflow for Mac that preserves your focus — no context-switching, no cloud uploads.

Tucky is voice to text for your whole Mac. Press a hotkey, speak, and polished text appears wherever your cursor is. We spend our days prompting AI now, and prompts don’t need perfect wording. They need your thinking, fast. Speaking is the fastest way to get it out.
Each capture method feeds the next. First your voice replaces the keyboard. Then your meetings and screen recordings join in. Everything lands in one searchable place — and once it’s all there, Tucky can tell you what happened, what matters, and what to do next.
See how the loop worksPress a hotkey and talk. Clean text lands in any app: AI chats, email, Slack, code.
Record Zoom calls with live transcription and call guidance, and narrate screen recordings for clients.
Choose project tags yourself or enable automatic suggestions, then check the result.
Daily recaps, chat with your history, export to ChatGPT, and see what to automate next.
Voice to text is the fastest way to work with AI. You don’t need exact wording to prompt ChatGPT or Claude — you need the idea out of your head. Press a hotkey, speak naturally, and Tucky puts clean, polished text at your cursor in any app on your Mac.
Start a dictation with your trigger word and the AI takes instructions. Say “echo, format this as an email” and your stream of thought comes out as a clean email in your tone of voice. Formats, terminology, and the trigger word itself are fully customizable — and plain dictation stays instant, because AI only steps in when you call it.
“Echo, format this as an email… hey Sarah quick one, sprint review moved to Thursday 3pm, can you update the calendar and double-check the demo build works”
Hi Sarah,
Quick heads-up: the sprint review has moved to Thursday at 3 pm. Could you update the calendar invite and double-check that the demo build is working?
Thanks!
Tucky notices when you join a Zoom, Meet, or Teams call and records it if you want it to. You get a live transcript as the call happens. Set up a template for what a great call looks like — a sales call, a discovery call, an interview — and Tucky tracks what’s been covered and suggests the next question in real time. None of your ideas get missed.

Record a narrated product demo, client walkthrough, or bug report. Edit the recording locally, then export it or upload it to Google Drive. Generate a transcript when you want to search the narration.

Keep dictations, notes, tasks, and meeting transcripts in a searchable local library. Project tagging is optional and can be wrong when a capture lacks context. Name the project in a voice note and review its assignment.

Ask questions about your saved work and inspect the sources behind each answer. Optional daily recaps are still being improved. For deeper prompt or workflow reviews, export notes or connect a compatible AI tool through MCP; that tool performs its own analysis.
One tool, a lot of jobs. Here’s what Tucky ends up doing for people.
Talk your prompts into ChatGPT, Claude, or Cursor. Rough wording is fine — speed is the point.
Customizable meeting guides, post-call feedback, and structured notes to review and export to your sales tools.
Screen record proposals, demos, and handoffs instead of scheduling another meeting.
Every call transcribed and summarized. Find any decision in seconds, weeks later.
Every idea captured and searchable. Chat with your history and get answers, not files.
Use exports or an MCP-connected AI tool to review repeated prompts and identify workflow changes you can make.
Tucky recognises the command and keeps your computer awake for the requested duration. Use it while a long-running command finishes or a queue of AI prompts works through its next steps.
Press your usual dictation hotkey and say:
“Echo, keep my computer awake for two hours.”No cloud backend. No account. No data sold. Dictation, recordings, transcripts, and project context stay under your control on your Mac.
Dictation uses your hotkey. For supported meeting apps, choose always record, ask first, or never. Screen recording starts when you launch it.
Core transcription and local AI work offline after model downloads. Google Drive uploads, downloads, updates, and external AI services need a connection.
Core app features need no Tucky account. Optional Google Drive sharing requires connecting your Google account.
Notes live in a SQLite file on your Mac. Export it, back it up, or delete it — full control, always.

Stop losing decisions between back-to-back meetings. A practical capture workflow for Mac that preserves your focus — no context-switching, no cloud uploads.

Turn your meetings, voice notes, and screen recordings into a searchable local knowledge base on your Mac. Everything stays private — no cloud, no uploads.

Stop typing every prompt. One hotkey lets you speak into ChatGPT, Claude, Cursor, and any app — all on your Mac, all private, no cloud uploads.
Open Terminal and paste the line below. No Homebrew required.
curl -fsSL https://raw.githubusercontent.com/desduvauchelle/echo-scribe/main/install.sh | bash