Speak. Get text.
No cloud required.
Limato Voice transcribes your speech directly in Chrome using an on-device AI model. Your audio never leaves your browser — not even for a millisecond.
Free tier: 20 transcriptions/day · Pro: $9.99 one-time
What it does
Fully offline
After the one-time model download, everything runs inside your browser with zero network requests.
Works everywhere
Gmail, Notion, Google Docs, LinkedIn — dictate into any text field on any page.
Auto-language detection
Speak in any language. The model detects it automatically, no configuration needed.
Inserts at cursor
Text is inserted directly where you're typing — no copy-paste needed, React and Vue inputs included.
Your audio never leaves your device
Unlike cloud-based dictation tools, Limato Voice runs Whisper speech-recognition models locally via WebGPU or WASM. No audio is uploaded to any server — ever. The only network traffic is a one-time model download and optional license activation.
How it works
Install & download model
Add the extension, then download the on-device model once (140–500 MB depending on model). Takes a few minutes on first launch.
Press ⌥ Space (Alt+Space)
Press the shortcut anywhere in Chrome to start recording. Or click the Limato Voice icon and press the mic button.
Text appears
Stop recording — your speech is transcribed and inserted into the focused field on the page.
Pricing
Free
$0 forever
Get started at no cost.
- 20 transcriptions per day
- All languages
- Full offline processing
Pro
$9.99 one-time
Pay once, use forever. No subscription.
- Unlimited transcriptions
- All languages
- Full offline processing
- 14-day money-back guarantee
From the LIMATO family
Need to polish your writing too?
Try LIMATO — an AI rewriter and translator that turns rough drafts into native-quality English in one click.
Cloud-based (uses an LLM API). Limato Voice stays 100% local — always.