What it is
OpenWhispr is an open-source desktop dictation app from the OpenWhispr organization on GitHub. Press a global hotkey, speak, and the transcribed text is pasted at your cursor in whatever app you are using. It can also translate as you dictate, transcribe meetings with speaker identification, transcribe audio and video files, and save everything as notes with meaning-based search. Speech recognition can run on your own computer with models such as Whisper and NVIDIA Parakeet, which the documentation says keeps your audio on your device, or you can switch to cloud models for speed. The AI assistant can connect to GPT, Claude, Gemini, Groq, or local models.
Who it's for:people who write, take notes, or sit in a lot of meetings and want to type less. You install it from a download, no coding needed.
Before you start: available for macOS (Apple silicon and Intel), Windows, and Linux, with .dmg, .exe, .AppImage, .deb, and .rpm packages. The documentation says the app is free and open source, and team spaces and cloud sync are free for signed-in users. Cloud models need an account with the provider, billed by that provider. On Intel Macs, live speaker identification and voice fingerprinting are unavailable, while meeting transcription and notes still work. MIT license.
Based on the project's public documentation. Not tested by AI Inspiration World.