Turn long recordings into clean text.
Cue transcribes while you record — and then does the next step. Lectures, interviews, and voice memos become clean, punctuated text you can search, summarize, translate, or turn straight into the follow-up work.
Free download · macOS & Windows · No credit card
…so the deadline moves to Thursday, and design review stays on Monday.
Two open questions: pricing page copy, and who owns the launch email.
Let's get the draft out today so legal has time to look at it.
— transcribing live on your desktop
44% lower dictation latency, zero marginal inference cost
Featured by Google DeepMind — Gemma 4 running on-device
What people transcribe with Cue
Lectures & talks
Record an hour-long lecture and leave with a transcript you can question: "what were the three arguments?"
Interviews & research
Stay in the conversation instead of typing. The transcript lands as a document you can quote and summarize.
Voice memos & drafts
Think out loud on a walk. Cue turns rambling audio into structured notes you can actually reuse.
Recording meetings? See meeting notes without a bot →
Live transcript, not an upload queue
Most audio-to-text converters make you record somewhere else, export a file, upload it, and wait. Cue transcribes on your desktop while you speak — the text streams in live, and the recording stays with your notes.
When the session ends, the transcript is a starting point, not a dead end. Ask Cue to summarize it, pull out decisions, translate a section, or draft the follow-up email — in the same window, without copying text between tools.
Fast because it runs on your machine
Google DeepMind featured Cue in its official Gemmaverse for exactly this architecture: running the Gemma 4 model on-device cut dictation latency by 44%, lifted feature usage by 30%, and dropped marginal inference cost to zero. Local speed is also why Cue can afford a genuinely useful free tier.
Voice transcription FAQ
Is Cue voice transcription free?
Cue is free to start: download the app, hit record, and get a live transcript on your desktop. Longer sessions and heavier AI processing are covered by the free daily allowance, with paid plans for more.
How long can a recording be?
Cue is built for long-form sessions such as lectures, interviews, and working sessions. The transcript streams live while you record, so you are not waiting for a file to upload afterwards.
What is the difference between voice typing and transcription?
Voice typing inserts short dictated text into the app you are using. Transcription captures a long recording as a document you can search, summarize, and reuse. Cue does both from the same hotkey.
Which languages does Cue transcribe?
Cue supports 20+ voice input languages, including English, Chinese, Japanese, Korean, Spanish, French, and German.
They hear what you said.
Cue sees what you're doing.
And does the thing, in any app.