Resonate

Voice to Notes, Instantly

Record conversations, meetings, and ideas.
Resonate transcribes your audio into clear, structured notes and key insights.

---

Features

• Record - Capture microphone and system audio simultaneously
• Transcribe - Local speech-to-text via bundled Whisper (offline, private)
• Summarize - AI-powered meeting notes, action items, and summaries via configurable LLM endpoints (OpenAI, Anthropic, or custom)
• Organize - Notebooks, drag-and-drop, tabs, and workspace cards
• Action Items - Auto-extracted tasks with assignees, checkboxes, and markdown
• Speaker Detection - Automatic speaker turn detection with renameable labels
• Export - Download transcripts, summaries, and action items as Markdown
• Mini Window - Pop-out recording widget with always-on-top pin
• Dark & Light Themes - Brand-consistent UI with Resonate color palette
• Tutorial - Built-in interactive walkthrough with example data

Quick Start

Prerequisites

• Node.js 20+
• Git

Install & Run

git clone https://github.com/ghost-ng/Resonate.git
cd Resonate
npm install
npm start

Note: This repo uses Git LFS for the Whisper model file (~142 MB). Install Git LFS before cloning: git lfs install

Whisper binaries and the base.en model are bundled in the repo under resources/whisper/ — no downloads needed. Works fully offline.

AI Configuration

Configure your AI endpoint in Settings > AI Endpoint:

Provider	Endpoint	Notes
OpenAI	https://api.openai.com	Requires API key
Anthropic	https://api.anthropic.com	Requires API key
Local (Ollama, LM Studio)	http://localhost:11434/v1	No key needed

Building

Development

npm start          # Start in dev mode with hot reload
npm run lint       # Run ESLint
npm test           # Run tests

Package for Windows x64

# Package (creates portable app in out/)
npm run package -- --arch=x64 --platform=win32

# Build installer (creates Squirrel installer)
npm run make -- --arch=x64 --platform=win32

The packaged app will be in out/Resonate-win32-x64/.

Native Dependencies

This project uses better-sqlite3 and native-recorder-nodejs which require native compilation. If you encounter build issues:

# Rebuild native modules for Electron
npx electron-rebuild

Packaging Checklist

All external dependencies are bundled in the repo:

1. Whisper binaries and model are in resources/whisper/ (tracked via Git LFS)
2. resources/whisper/ is bundled into the app via extraResource in forge config
3. Native modules (better-sqlite3, native-recorder-nodejs) are rebuilt for Electron
4. Run npm run make to create the installer — works fully offline

Project Structure

src/
  main/              # Electron main process
    db/              # SQLite database, migrations, repositories
    ipc/             # IPC handlers
    services/        # Audio capture, STT, AI summary, process monitor
    stt/             # Speech-to-text engines (Whisper)
  preload/           # Electron preload script
  renderer/          # React UI
    components/      # UI components
    stores/          # Zustand state stores
    hooks/           # React hooks
    lib/             # Utilities and formatters
  shared/            # Types and constants shared between processes

Tech Stack

Layer	Technology
Framework	Electron 41
Frontend	React 19, TypeScript, Tailwind CSS 4
State	Zustand
Database	SQLite (better-sqlite3)
STT	whisper.cpp (local)
AI	OpenAI / Anthropic API (configurable)
Build	Electron Forge, Vite
Audio	native-recorder-nodejs

License

MIT