doing.

Voice transcription for AI builders

Talk to Claude,
don't type.

Fast local voice transcription, no subscription.

macOS 14+ · Apple Silicon · No account required

Hold fn Talk. Done.

1Hold your hotkey

Doing starts listening

2Speak

Voice transcribed in real-time

3Release to paste

Text is pasted at your cursor

Live pip follows your mouse.

A small pip follows your cursor so you always know where your words will land.

Want to go faster? Turn on YOLO Mode.

Your voice never leaves your Mac.

No audio uploaded. No cloud transcription. No account. Everything runs on your machine.

This isn't a privacy policy. It's architecture.

Own your tools, stop renting.

Other voice tools charge $8–15 a month, forever. Doing is $49 once. It's yours. No recurring charges, no subscription anxiety. Just a quality tool that you own and can use forever.

doing

$49 once

Year 2 cost: $0

Runs locally

No account needed

Yours forever

Other voice tools

$8–15/mo

$96–180 per year

Cloud-dependent

Account required

Cancel and it's gone

3 device activations · 14-day refund guarantee · Free updates

Transcription with zero wait.

Doing transcribes locally in real-time — no upload, no server, no spinner. Your words are already there.

150xrealtime

60 seconds of audio transcribed in ~400 milliseconds. On-device.

Parakeetlocal
150x
Applelocal
25x
Geminicloud
23x
OpenAIcloud
22x
AssemblyAIcloud
20x

Measured using doing's built-in benchmark tool. Cloud providers available via bring-your-own-key. Powered by Parakeet-TDT by NVIDIA, licensed under CC-BY-4.0.

Level up with YOLO Mode.

Turn it on and Doing automatically presses Return after pasting your transcription. Speak, release, and your prompt is already running.

YOLO Mode
Speak
Review transcript
Edit / fix mistakes
Press Return
Done

YOLO is a mindset shift. Stop editing, stop polishing, just talk. LLMs know what you mean, even when your words aren't perfect.

Works with every AI tool and app.

Doing works at the system level. Anywhere you can type, you can talk. Hold fn in your browser, editor, or terminal — Doing transcribes and pastes wherever your cursor is.

✳ Claude Code
myappgit:(main)claude

Claude Code v2.1.45

Opus 4.6 (1M context) · ~/doing/myapp

myapp on main | Opus 4.6 (1M context)
Claude CodeChatGPTClaude.aiCursorCodexWindsurfSlackNotionGmailAny text field

Audio Ducking

Keep your music playing.

Doing automatically fades your audio down when you start recording and brings it right back when you stop. No fumbling for pause. No breaking your flow.

Music playing

Post-processing

Customize with Skills.

Write prompts that automatically process your transcription before it's pasted. Clean up filler words for Slack. Formalize for email. Convert to a code prompt for Cursor.

Skills are app-aware — set different skills for different apps and they trigger automatically based on where you're pasting.

Runs on-device with Apple Intelligence, or plug in your own API key for GPT-4o, Claude, or any OpenAI-compatible endpoint.

Manage Skills
FormalizeGmail

Rewrite in a professional tone with proper punctuation and structure.

Code PromptCursor

Convert spoken description into a clear, technical instruction for an AI coding assistant.

SummarizeNotion

Condense into concise bullet points. Keep key details, drop filler.

EmojifyMessages

Replace the entire transcript with emoji. No text, just emoji.

4 skills+ New Skill

5 engines. 1 benchmark tool.

Choose your engine with data, not marketing.

Doing ships with Parakeet — an on-device model that transcribes 15 seconds of audio in 180 milliseconds. But we don't lock you in. Bring your own API key and benchmark every engine side-by-side on the same audio.

Local · free forever: Parakeet (v3), Apple Foundation

Cloud · bring your key: OpenAI Whisper, Gemini, AssemblyAI

Cloud engines use your own API key — you pay the provider directly, per use. No middleman markup.

Benchmark Results
15.0s audio5 providers
Parakeet (v3)localactivefastest
Time0.10s
Speed150x realtime
Costfree

Okay, we're gonna test the benchmarking tool. I'm just speaking out loud here while um to collect some words for the transcription process.

Download doing to run your own benchmarks

Built by a builder

Brian and Rosie

I built the voice tool I wanted.

I'm Brian. I use AI tools like Claude Code every day, issuing hundreds of prompts. I got tired of typing all of them — and I didn't love that every voice tool on the market was sending my audio to their servers.

So I built doing. Now it's how I build everything — including doing itself.

Doing is bootstrapped. No investors, no pressure to become a platform or monetize your data. Just one person building a tool that does one thing well.

Local. Fast. Yours.

Voice input for AI builders. No cloud. No subscription. $49 once.

Try it free for 100 transcriptions — no account, no credit card.

macOS 14+ · Apple Silicon · No account required