Cloudless Voice

Cloudless Voice is fast voice dictation that runs on your device. Speech recognition happens locally, so your voice never leaves your computer or phone, it works offline, and your data is never used to train remote AI models.

Choose your model Pick the speech model that works best for you, including new open models as they're released. Correct your past transcripts, and uses them to backtest models against your own voice and language, so you can see which one is most accurate for you before you switch.

Clean up your way

• Bring your own key (BYOK): connect your own AI provider or local model to clean up transcripts
• Or upgrade to Smart Cleanup ($5.99/month): our custom-built local AI that polishes every transcript on-device before it's pasted

Smart Cleanup:

• Removes filler words ("um," "uh," "like," "you know")
• Handles self-corrections ("9 am, no sorry, 10 am" → "10 AM")
• Adds punctuation and capitalization
• Formats numbers, times, email addresses and lists

Features:

• Works in any app or text field
• Hold a hotkey to dictate (Caps Lock by default, customizable), or double-tap for hands-free mode
• Custom dictionary that learns your vocabulary
• Snippets: speak a phrase to insert saved text
• Transcript history and personal stats like words per minute
• 100+ languages
• Light and dark mode

Pricing: the app is free. Smart Cleanup is $5.99/month with unlimited words, less than half the price of cloud dictation apps like Wispr Flow ($12–15/month).

For teams, the Enterprise plan brings private, on-device dictation to organizations that can't send voice data to third-party servers.