PenSistant

Speak naturally, write effortlessly.

AI-powered voice-to-text dictation that works anywhere on your Mac. Record with a hotkey, let AI transcribe and format, and watch your words appear at the cursor.

Download on the Mac App Store

Everything you need,
nothing you don't.

Designed to disappear into your workflow. PenSistant stays in the menu bar until you need it, then gets out of your way.

Speech to Text

Speak naturally and let PenSistant turn your voice into polished, formatted text — inserted right where your cursor is.

Voice Recording

Trigger recording with a global hotkey from any application. No window switching needed.

AI Transcription

Powered by OpenAI Whisper for accurate, multi-language speech recognition.

Smart Formatting

GPT-4o-mini polishes grammar, punctuation, and structure automatically.

Walkietalky Mode

Hold Right Command, speak, release. Instant Whisper-only transcription — no formatting delay.

Multi-language

Automatic detection and transcription for Portuguese and English out of the box.

Stacking Notes

Each transcription opens its own window in the lower corner of your screen. As you dictate more, windows stack and cascade on top of each other — building a history of notes you can always go back to. Need something from earlier? Just click through the stack. Every window stays fully alive: edit the text, polish it, format it, copy it, or close it independently.

Select to Voice

Highlight any text on your screen and let PenSistant read it aloud. Powered by OpenAI's text-to-speech API with real-time streaming — audio starts playing before the full response arrives.

Select & Speak

Highlight any text in any app and press Opt+Cmd+R. PenSistant grabs the selection and starts reading it aloud instantly.

Streaming Playback

Audio begins playing while the rest is still being generated. No waiting for the full file — hear results in under a second.

2x

Speed Controls

Adjust playback speed on the fly — 0.5x for careful listening, up to 2x for quick reviews. Play, pause, and scrub through audio.

Privacy First

API key secured in macOS Keychain. No telemetry, no data collection, no cloud storage.

Menu Bar App

Lives in your menu bar — lightweight, instant access, never in the way. Under 20MB.

Simple by design.

Two powerful workflows, each just a hotkey away.

Speech to Text

⌥⌘T

Press the Hotkey

Hit Option+Command+T from anywhere on your Mac. PenSistant captures your cursor position and starts recording.

Speak Naturally

Talk as you normally would — in English or Portuguese. Whisper handles accents, pace, and pauses intelligently.

Text Appears

Your words — transcribed, formatted, and polished — are inserted right where your cursor was. Like magic.

Select to Voice

⌥⌘R

Select Text

Highlight any text on your screen — an article, email, document, code comments — and press Option+Command+R.

Audio Streams

OpenAI's TTS generates speech in real time. Audio starts playing immediately — no waiting for the full file to download.

Listen & Control

A playback window appears with play, pause, progress scrubbing, and speed controls from 0.5x to 2x. Your text is shown alongside the audio.

Get started in minutes.

PenSistant uses your own OpenAI API key. Here's how to set it up.

1

Create an OpenAI account

Go to platform.openai.com and sign up for an account if you don't have one already.

2

Generate an API key

Navigate to API Keys in your OpenAI dashboard. Click Create new secret key, give it a name like "PenSistant", and copy the key. You'll only see it once.

3

Add billing credits

Go to Billing and add a payment method. The Whisper and TTS APIs are pay-per-use — typical dictation usage costs just a few cents per day.

4

Paste the key into PenSistant

Open PenSistant, go to Settings (gear icon in the Home window or via the menu bar), and paste your API key. PenSistant validates the key automatically and stores it securely in your macOS Keychain — never in a file.

5

Grant Accessibility permissions

macOS will prompt you to allow PenSistant in System Settings → Privacy & Security → Accessibility. This is required to capture your cursor position and insert text via keyboard simulation. Microphone access is also needed for recording.

Your API key never leaves your machine — it's stored in the macOS Keychain and sent directly to OpenAI's servers over HTTPS. PenSistant has no backend, no accounts, and no telemetry.

Built by hand.

Andre Arruda logo

Andre Arruda

Head of Engineering

Forward-thinking engineer with a strong record of driving business growth through technology strategy and execution. With expertise spanning product engineering, UI/UX design, AI integrations, and agile methodology, PenSistant was born from a personal need — dictating text anywhere on the Mac without breaking flow. Stanford AI-Driven Leadership certified.

Support PenSistant

PenSistant is open source and free to use. If it saves you time, consider buying me a coffee to keep development going.

Every contribution helps fund hosting, API testing, and new features.