A privacy-first meeting transcription and summarization app that runs entirely in your browser. No data ever leaves your device.
- Live Transcription — Real-time speech-to-text using the Web Speech API
- Whisper Retranscription — Reprocess recorded meeting audio with local Whisper (
transformers.js) for cleaner transcripts - AI Summaries — Generate structured summaries (key points, decisions, action items) with Chrome's built-in Summarizer API
- Meeting Chat — Ask questions about any recorded meeting using the Prompt API
- Notes — Take notes alongside the transcript; edit later and regenerate summaries
- Editable Meeting Titles — Rename meetings directly from the detail page
- Audio Tab + Playback — Recorded audio is saved locally and can be played back per meeting
- Pause / Resume — Pause and resume recording as needed, with autosave every 10 seconds
- Transcribe More — Continue recording on an existing meeting and append new segments
- Settings Page — Configure default language and Whisper backend, view storage stats, and clear cached audio/model data
- Offline Storage — All meetings stored locally in IndexedDB
- Next.js 16 + React 19
- Tailwind CSS 4
- Web Speech API for transcription
- transformers.js + local Whisper ONNX model for retranscription
- Chrome Summarizer API for AI summaries
- Chrome Prompt API for meeting chat (feature-flagged)
- IndexedDB for local persistence
- Chrome or Edge (required for Speech Recognition and Chrome AI APIs)
- Gemini Nano enabled — go to
chrome://flags/#optimization-guide-on-device-modeland set to Enabled - Summarizer API enabled — go to
chrome://flags/#summarization-api-for-gemini-nanoand set to Enabled - Node.js 18+
pnpm install
pnpm devOpen http://localhost:3000 in Chrome.
- Record — Start a new meeting, speak, and the transcript appears in real time
- Note — Add notes in the side panel during or after the meeting
- Summarize — Generate an AI summary from the transcript and your notes
- Chat — Ask follow-up questions about the meeting content
- Retranscribe (Optional) — Run Whisper on saved audio for improved transcript quality
- Review — Browse past meetings from the home page and manage defaults from Settings
Everything runs client-side. Speech recognition, AI summarization, and storage all happen in your browser. No server, no third-party APIs, no data exfiltration.
