OpenWhispr turns speech into text anywhere on your computer. Press a hotkey, speak, and the words appear at your cursor. It can transcribe fully offline with local speech models, such as Whisper and NVIDIA Parakeet, or use cloud models with your own API keys.

Main features

• Dictation in any app with automatic pasting, plus a hotkey for dictating in one language and pasting in another.
• Voice-driven AI assistant: speak commands to models such as GPT, Claude, Gemini or local ones.
• Meeting transcription: detects Zoom, Teams and FaceTime calls, labels speakers and connects with Google, Microsoft or Apple calendars.
• Notes: folders, semantic search, optional cloud sync and AI actions.
• Audio import: transcribe existing audio or video files, including from a link.

Platforms and privacy

Available for macOS (Apple Silicon and Intel), Windows and Linux. The project states that it collects no data and has no telemetry. A few features, such as live speaker identification, are not available on Intel Macs. It is built with Electron, React and TypeScript.