Open-source, local-first meeting and voice memo recorder that transcribes on-device into Markdown files that AI agents like Claude Code can search via MCP.
1.6K+13 this week182RustMITWell maintainedUpdated 20 hours ago
- Stars
- 10,877
- Forks
- 1,421
- Open issues
- 509
- Last commit
- 23 hours ago
What is it?
OpenWhispr is a desktop app for macOS, Windows, and Linux. Press a global hotkey, speak, and the transcribed text is pasted into whichever application has the cursor. Transcription can run fully offline with local models such as Whisper, NVIDIA Parakeet, and Cohere Transcribe, so audio stays on the device, or it can use cloud providers for speed. Users can bring their own API keys for services like OpenAI, Anthropic, Gemini, and Groq.
Beyond dictation, the app includes a voice assistant hotkey that sends spoken commands to an AI model and can edit highlighted text in place. It also handles meeting transcription with calendar integration and speaker labelling, transcribes imported audio and video files, and offers a notes feature with folders and search. A public API and an MCP server allow other tools and AI assistants to access notes and transcriptions.
The desktop app is built with Electron, React, and TypeScript. Its Expo-based mobile app lives in a separate subdirectory with its own build setup.
Key features
- 01Global hotkey dictation that pastes text into any app
- 02Offline transcription with Whisper, NVIDIA Parakeet, and other local models
- 03Cloud model support via bring-your-own API keys
- 04Voice assistant hotkey for editing selected text with AI
- 05Meeting transcription with speaker diarization and calendar integration
- 06Public API and MCP server for programmatic access
Use cases
- Dictating emails, documents, or chat messages hands-free
- Transcribing meetings with speaker labels and searchable notes
- Converting recorded audio or video files and YouTube links to text
- Running speech recognition offline for privacy-sensitive work
Quick start
- 1
Clone the repository
git clone https://github.com/OpenWhispr/openwhispr.git - 2
Enter the project folder
cd openwhispr - 3
Install dependencies (requires Node.js 24+)
npm install - 4
Run the desktop app in development mode
npm run dev
Things to know
- Requires Node.js 24+ to build and run from source.
- On Intel Macs, live speaker identification and voice fingerprinting are unavailable.
In collections
Topics
FAQ
- What is openwhispr?
- Open-source desktop dictation app that turns speech into text at your cursor, with local Whisper/Parakeet models or cloud providers. OpenWhispr is a desktop app for macOS, Windows, and Linux. Press a global hotkey, speak, and the transcribed text is pasted into whichever application has the cursor. Transcription can run fully offline with local models such as Whisper, NVIDIA Parakeet, and Cohere Transcribe, so audio stays on the device, or it can use cloud providers for speed. Users can bring their own API keys for services like OpenAI, Anthropic, Gemini, and Groq.
- Is openwhispr free to use? What license does it have?
- openwhispr is open source under the MIT license. Check the license terms before using it in commercial products.
- Is openwhispr still maintained?
- Actively maintained. The last commit to openwhispr was on Nov 6, 2026.
- What can I use openwhispr for?
- Dictating emails, documents, or chat messages hands-free Transcribing meetings with speaker labels and searchable notes Converting recorded audio or video files and YouTube links to text Running speech recognition offline for privacy-sensitive work
- How do I install openwhispr?
- Clone the repository: git clone https://github.com/OpenWhispr/openwhispr.git Enter the project folder: cd openwhispr Install dependencies (requires Node.js 24+): npm install Run the desktop app in development mode: npm run dev
- Where can I get openwhispr?
- The source code and installation instructions are on GitHub: https://github.com/OpenWhispr/openwhispr
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source desktop AI note taker that transcribes meetings and voice notes without a bot joining the call, using local or cloud AI.
129+3 this week19TypeScriptMITUpdated 12 days ago
Open-source, local-first AI dictation app that turns speech into text on macOS and Windows, working offline with Whisper.
1.6K+20 this week144TypeScriptMITWell maintainedUpdated yesterday
Open-source macOS meeting-notes app that records mic and system audio without a bot, transcribes locally or via cloud APIs, and writes summaries from your notes plus the transcript.
33845RustMITUpdated 6 days ago
A macOS meeting note-taker that transcribes both sides of a call locally and surfaces talking points from your own notes.
2.6K+10 this week260SwiftMITWell maintainedUpdated 4 days ago
Privacy-first AI meeting assistant that records, transcribes and summarizes meetings locally on macOS and Windows, with optional Ollama or cloud LLM summaries.
32.5K+167 this week3.6KRustMITUpdated 24 hours ago