HN user
A macOS app that transcribes meetings in real time and generates smart summaries — fully offline, no API keys, no data leaves the machine. Captures mic and system audio (Zoom, Teams, Meet) via ScreenCaptureKit, runs Whisper Turbo for STT and Ministral 3B for summaries/chat, all on Apple Silicon. Open source, MIT.
oupsi ... wrong paste ...
MCP server that wraps AzerothCore's SOAP interface into 28 GM tools. Talk to your WoW private server through any LLM client instead of typing console commands — bans, accounts, mail, guilds, teleports, etc.
https://github.com/Lingelo/azerothcore-wotlk-mcp
If you find it useful, a star on the repo goes a long way!
I built a macOS app that transcribes meetings in real-time and generates notes (summary, action items, key points) using AI that runs entirely on your machine. No cloud, no API keys, no
subscription.
It captures both your microphone and system audio (Teams, Meet, Zoom) simultaneously using a native Rust module with ScreenCaptureKit, then runs Parakeet TDT (NVIDIA, 640MB) for speech-to-text and
Mistral 7B for summarization — all local via sherpa-onnx and node-llama-cpp.
Currently French only — multi-language support is on the roadmap. Pull requests are welcome if you want to help add other languages or contribute in any way.
Stack: Electron 34, Angular 21, NestJS 11, SQLite FTS5, Rust/napi-rs. Apple Silicon only (M1+, 16GB RAM minimum).
The main technical challenges were audio capture (resampling, multi-source mixing, format conversion) and keeping the UI responsive while running two AI models — solved by isolating STT and LLM in
separate Node.js workers.
MIT licensed. DMG available for download, no code signing yet so you'll need xattr -cr. I built a macOS app that transcribes meetings in real-time and generates notes (summary, action items, key points) using AI that runs entirely on your machine. No cloud, no API keys, no
subscription.
It captures both your microphone and system audio (Teams, Meet, Zoom) simultaneously using a native Rust module with ScreenCaptureKit, then runs Parakeet TDT (NVIDIA, 640MB) for speech-to-text and
Mistral 7B for summarization — all local via sherpa-onnx and node-llama-cpp.
Currently French only — multi-language support is on the roadmap. Pull requests are welcome if you want to help add other languages or contribute in any way.
Stack: Electron 34, Angular 21, NestJS 11, SQLite FTS5, Rust/napi-rs. Apple Silicon only (M1+, 16GB RAM minimum).
The main technical challenges were audio capture (resampling, multi-source mixing, format conversion) and keeping the UI responsive while running two AI models — solved by isolating STT and LLM in
separate Node.js workers.
MIT licensed. DMG available for download, no code signing yet so you'll need xattr -cr.In 2011, Notch created "Prelude of the Chambered" during Ludum Dare 21 – a Wolfenstein-style raycasting dungeon crawler in Java. I rewrote it entirely in TypeScript with Vite, using only Canvas 2D (no WebGL). Runs on any browser, no install needed. The original level design system where each PNG pixel defines a game element is preserved and quite elegant.