OpenWhispr

About

OpenWhispr is an open-source, free voice-to-text dictation app that turns your speech into text at your cursor. Press a global hotkey, speak, and your words are pasted into whatever app you're using. You can choose fully offline transcription with local speech-to-text models such as Whisper and NVIDIA Parakeet, so audio never leaves your device, or use cloud processing for speed. The app is described as having no data collection or telemetry, and its source code is publicly available.

Beyond dictation, OpenWhispr offers a dedicated hotkey for dictating in one language and pasting in another, and a Voice Assistant hotkey that sends spoken commands to an AI assistant such as GPT-5, Claude, Gemini, or local models. It can also transcribe meetings from apps like Zoom, Teams, and FaceTime with speaker labelling, offer a notes feature with folders and search, and import existing audio or video files.

OpenWhispr is available for macOS, with separate downloads for Apple Silicon and Intel Macs, and also runs on Windows and Linux. The website notes that on Intel Macs, live speaker identification and voice fingerprinting are unavailable, though meetings still record and transcribe normally. The app can also be installed via a community-maintained Homebrew cask.