Think
out loud.

Voice-to-text that works in Slack VS Code Gmail Notion ChatGPT every app

Press a hotkey, speak, release — your polished text is in the clipboard.

Free & open source · Bring your own API key

3.75x

faster than typing

The average person types 40 words per minute. The average person speaks 150. Waffler bridges the gap.

Three steps. Sixty seconds. Done.

Step 1

Press and hold your hotkey

Hit your global shortcut from any app to start recording

Step 2

Speak naturally

Talk like you normally would — say whatever's on your mind

Step 3

Release — text in clipboard

Let go of the hotkey and polished text is ready to paste

What happens when typing gets out of the way

Voice input should be effortless. Waffler handles transcription, cleanup, and pasting — you just speak.

Global Hotkey

Press your hotkey from any app. Waffler activates instantly — no window switching needed.

OpenAI Whisper or Groq

Choose your transcription provider. Both are accurate. Groq is faster and often cheaper — set whichever API key you have.

AI Cleanup

Filler words removed, grammar fixed automatically — so your text reads like you typed it deliberately.

Auto-Copies to Clipboard

Cleaned text lands in your clipboard the moment transcription is done. Paste it anywhere.

Local Transcription History

Every transcription is saved on your device and is searchable. Nothing leaves your machine except the audio you send to your API provider.

Mac + Windows · No Account

Native installers for macOS and Windows. Download, add your API key, and you're set — no account or sign-up required.

We never hear you. By design.

Privacy isn't a feature — it's how Waffler is built.

Audio goes directly to OpenAI or Groq

Audio travels from your machine to OpenAI or Groq under your own API key — never through any Waffler servers.

History stays on your machine

Transcription history is stored locally on your device only. It never leaves your machine.

No analytics, no telemetry

Waffler collects nothing. No usage data, no crash reporting, no account required.

Frequently asked questions

Everything you need to know about Waffler. Can't find what you're looking for? Get in touch.

Do I need an account?
No. There is no Waffler account. Download the app, set your API key in the .env file, and you're done.
Is it really free?
Yes — Waffler itself is free and open source (MIT licensed). You pay OpenAI or Groq directly for API usage at their standard rates, roughly $0.006 per minute with OpenAI Whisper, often cheaper with Groq.
Is my data private?
Audio goes to OpenAI or Groq under your own API key — Waffler's servers are never in the loop. Your transcription history is stored locally on your machine only and never leaves your device.
What's the difference between OpenAI and Groq?
Both work well. Groq is generally faster and often cheaper. OpenAI Whisper is more widely used and well-documented. Set whichever API key you already have in the .env file — you can switch at any time.
Why does it warn about an unidentified developer / unsafe app?
Waffler builds are unsigned — code signing certificates cost money and this is a free indie project. On Mac, right-click the app and choose Open to bypass Gatekeeper. On Windows, click More Info then Run Anyway to get past SmartScreen. This is normal for unsigned open-source software.
Mac or Windows?
Both are supported. Separate installers are available on the GitHub releases page at github.com/jbf-tars/waffler/releases.

Ready to stop typing?

You've read enough. Time to talk. Download Waffler and start thinking out loud.

Free & open source · Mac + Windows · No account required