Local-first voice-to-text for macOS.
What it is
Hold Fn, speak, let go. Text appears where your cursor is.
All transcription happens on-device. Your audio lives in memory only while you hold the key. The moment it's transcribed, it's gone.
Text lands at your cursor, in Notion, Slack, VS Code, Mail, anywhere.
Switch between English, Tamil, Hindi mid-sentence. It keeps up.
Who it's for
If your first drafts are messy, personal, or just too much to type. Sollu is for you.
A pen's job is to write. Not to remember. Not to transmit.
A voice-to-text app, in the digital era, should act like a pen. It should hear you and write it down. Nothing more. There are already plenty of notepads, audio archives, and second brains. Those are different products, for different jobs.
I kept looking for that pen, and couldn't find one. I'm a writer who's lazy to type, and I don't want an LLM generating my first draft. Every voice-to-text app I tried had a trade-off I couldn't live with: the transcription wasn't good enough, the privacy was someone else's policy document, or my words were quietly being used to train a model I'd never see.
So I built sollu for myself. Since then I've been publishing at least one essay a week, brainstorming with Claude out loud instead of typing, and dumping thoughts as fast as they form. Friction, gone.
If that sounds useful to you too, I'd love to have you in the beta.
How it works
Hold Fn. Sollu captures audio into memory, not to disk.
An on-device model turns your voice into text. Fast on Apple Silicon.
Light cleanup. Punctuation, filler words. Your voice stays your voice.
Text lands at your cursor, in the app you're already in. Audio is gone.
Private beta
This is a small, hand-picked beta. I'm not looking for a thousand signups. I'm looking for a handful of people who will actually use it, tell me when it breaks, and help shape what it becomes.
macOS · Apple Silicon · ~2 min form · [email protected]