EdNoteAI is an AI-powered web application that helps students and professionals transform video and audio lectures into structured, academic-grade notes. The platform uses speech-to-text and language models to generate transcriptions, summaries, and key points.
- User authentication: Sign-up, login, and account management with Supabase Auth (password reset, change password, account deletion).
- File upload: Upload audio/video via the dashboard. Files are stored in AWS S3 with presigned URLs.
- Transcription: Uploaded media is transcribed by an AWS Lambda function using OpenAI Whisper.
- AI note generation: Transcripts are processed by a Lambda function (Google Gemini) to produce Markdown notes with LaTeX math.
- Real-time updates: The analysis page uses Supabase Realtime so the UI updates as soon as processing finishes.
- Media playback: Playback of uploaded audio and video in the browser.
- Notes display: Transcriptions and AI notes are rendered in Markdown with LaTeX (KaTeX). Summaries and key points are shown.
- Library: Searchable, filterable library of media and generated notes.
- Subscriptions & billing: Stripe integration for plans, checkout, billing portal, webhooks, and usage-based limits.
- Export: Export notes/transcripts in multiple formats (e.g. Markdown, plain text).
- Frontend: Next.js (App Router), React, TypeScript, Tailwind CSS,
react-markdown,react-katex - Backend: Next.js API routes, Supabase (PostgreSQL, Auth, Realtime), AWS Lambda, AWS S3, OpenAI API, Google Gemini API, Stripe, Resend
- Database: Relational schema for videos, transcripts, notes, library, subscriptions, and user settings, with RLS.
src/app/— Next.js app (pages, API routes, dashboard, analysis, upload, library, subscription, etc.)src/components/— Reusable UI (buttons, cards, modals, etc.)src/lib/— Services and utilities (Supabase, Stripe, subscription, email, auth)src/utils/— Supabase client helpers and shared utilitieschrome-extension/— Chrome extension (experimental; seechrome-extension/README.md)lambda_function-audio_trans.py— AWS Lambda for transcription (OpenAI Whisper)lambda_function-note_gen.py— AWS Lambda for note generation (Google Gemini)
Create a .env.local file (never commit it). Example variables:
Next.js app
| Variable | Purpose |
|---|---|
NEXT_PUBLIC_SUPABASE_URL |
Supabase project URL |
NEXT_PUBLIC_SUPABASE_ANON_KEY |
Supabase anon (public) key |
SUPABASE_SERVICE_ROLE_KEY |
Supabase service role key (server-only) |
NEXT_PUBLIC_APP_URL |
App URL (e.g. https://example.com) |
NEXT_PUBLIC_SITE_URL |
Site URL for emails/links |
STRIPE_SECRET_KEY |
Stripe secret key |
STRIPE_WEBHOOK_SECRET |
Stripe webhook signing secret |
RESEND_API_KEY |
Resend API key for transactional email |
REGION_AWS |
AWS region |
ACCESS_KEY_ID_AWS |
AWS access key ID |
SECRET_ACCESS_KEY_AWS |
AWS secret access key |
S3_BUCKET_NAME_AWS |
S3 bucket name |
TRANSCRIPTION_LAMBDA_FUNCTION_NAME_AWS |
Transcription Lambda name |
CLEANUP_SERVICE_KEY |
Optional; for batch S3 cleanup service |
Transcription Lambda (lambda_function-audio_trans.py)
SUPABASE_URL,SUPABASE_KEY— Supabase (service role)OPENAI_API_KEY— OpenAI API keyNOTE_GENERATOR_LAMBDA_ARN— ARN of the note-generation Lambda
Note-generation Lambda (lambda_function-note_gen.py)
SUPABASE_URL,SUPABASE_KEY— Supabase (service role)GEMINI_API_KEY— Google Gemini API key
- Clone the repo and install dependencies:
npm install
- Configure
.env.localwith the variables above. - Run the dev server:
npm run dev
- Open http://localhost:3000.
Deploy the Lambda functions to AWS with the appropriate env vars and wire S3/upload/transcription/note-generation as in the API routes.
A Chrome extension lives in chrome-extension/ for capturing tab audio and sending it to EdNoteAI. It is not fully functional at this time. See chrome-extension/README.md for details and development setup.
Contributions are welcome. Please open issues or pull requests for bugs, features, or improvements.