Wordbird: free dictation for Mac running Nvidia Parakeet locally
Dear HN, Wordbird is an open-source voice dictation app for macOS, powered by Nvidia Parakeet running locally on Apple Silicon via MLX. You can teach it project-specific terms like brand names or frameworks using a project-level WORDBIRD.md file. Run it via `uvx wordbird`. Press a hotkey, speak, and your words are transcribed and pasted into whatever app is focused. A small LLM, also running locally, post-processes the transcription to fix errors. If you use apps like VS Code, Zed, Terminal.app or iTerm, Wordbird will know which directory you're in, look up a project-specific WORDBIRD.md…
What it does
In the maker’s words, at launch
Dear HN, Wordbird is an open-source voice dictation app for macOS, powered by Nvidia Parakeet running locally on Apple Silicon via MLX. You can teach it project-specific terms like brand names or frameworks using a project-level WORDBIRD.md file. Run it via `uvx wordbird`. Press a hotkey, speak, and your words are transcribed and pasted into whatever app is focused. A small LLM, also running locally, post-processes the transcription to fix errors. If you use apps like VS Code, Zed, Terminal.app or iTerm, Wordbird will know which directory you're in, look up a project-specific WORDBIRD.md file, and if it exists, use it for post-processing the transcription. With the VS Code extension this even works via SSH. My friend Till Hoffmann wrote this because he was "tired of yet another subscription", and standard dictation just gets too many technical terms wrong. I'm posting this here on his behalf because he's not as active on HN. I've really fallen in love with Wordbird, already made some small PRs, and I hope that many here will find it equally useful. Really eager to hear feedback (I'll make sure Till monitors this thread as well)!
Does the same job
all alternatives →
TongueType for macOSMay 2026 · tonguetype.app · ▲109Local dictation for macOS without the subscription
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
