Skip to content
Garon edited this page Jun 13, 2026 · 2 revisions

Glimpse Wiki

On-device voice dictation. Open-source, local-first, private by default.

Core dictation is free and runs entirely on your machine. AI writing features are optional and only use your own LLM provider.

Download · Website · Privacy · Source


Getting started

  1. Download the latest release for macOS 14+ or Windows 10+.
  2. Open Glimpse and complete the one-question setup.
  3. Press your hotkey, speak, and your words land at your cursor.

Speech models live in Settings → Models. AI writing is in Settings → Providers.

What's free

  • Local transcription, offline, ANE-accelerated on Apple Silicon.
  • Custom dictionary for names, brands, and terms.
  • Auto Dictionary that picks up your words on its own.
  • Replacements: say "my address," get the full thing.
  • History and search.

Glimpse Personal

Included in the 14-day trial, then a one-time license:

  • Library: drop in audio or video, scrub the synced transcript, assign speakers, export to .txt, .md, .srt, .vtt.
  • AI Cleanup with your own LLM.
  • Edit Mode: highlight text, say what you want, watch it rewrite in place.
  • Personalization: different tones per app or site, with snippets.
  • Local API: an OpenAI-compatible speech endpoint on your machine.
  • CLI: an optional glimpse terminal command.

Guides

Privacy

Transcription stays on-device by default. Glimpse never collects your transcripts, audio, prompts, or API keys. Anonymous usage telemetry is tied to a random install ID, not your identity, and stored in the EU. See Analytics for the full breakdown. Opt out anytime in Settings → App.

If you enable an external LLM provider, text for Cleanup, Edit Mode, and Personalization is sent to that provider when those features run. Your API key stays local.

Help and contributing

Getting Started

Features

Help

Clone this wiki locally