Explainer AI

AI summarizer · podcast generator · quiz maker · mind map builder
Paste a URL, YouTube link, or PDF — get a podcast, summary, quiz, or mind map in seconds.

Overview

Explainer AI is an open-source, AI-powered learning platform that transforms web articles, YouTube videos, PDFs, and plain text into structured learning materials — summaries, podcasts, quizzes, notes, mind maps, and more.

Point it at any URL, paste a YouTube link, or upload a PDF and it generates ready-to-use content in seconds: an audio podcast you can listen to on the go, a structured summary, a quiz to test your understanding, or a visual mind map of the key concepts. Built for students, researchers, and anyone with more tabs open than time to read them.

The backend runs a decoupled Express API with two dedicated BullMQ workers (podcast + gossip audio) so long-running AI generation jobs never block the main server.

What can I use it for?

I want to…	Use this tool
Listen to a blog post or article during my commute	Podcast Generator
Get the key points from a YouTube video without watching it	Summarizer (YouTube link)
Turn a research PDF into an audio summary	Podcast Generator (PDF upload)
Test my understanding of a topic	Quiz Generator
Get structured notes from a web page	Smart Notes
Visualize how concepts in an article relate to each other	Visualizer (mind map)
Ask questions about something I just read	Chat
Deep-dive into a concept I didn't fully understand	Deep Explain

Features

• Summarizer — Condenses web pages, YouTube videos, PDFs, and pasted text into clear summaries with adjustable depth.
• Podcast Generator — Produces a full audio podcast from any content using AI-generated scripts and multi-provider TTS (Gemini TTS, ElevenLabs, Deepgram).
• Gossip Mode — Generates a casual, conversational audio episode styled as a gossip podcast — a second voice format with its own queue worker.
• Quiz Generator — Creates interactive multiple-choice and open-ended quizzes from source content.
• Smart Notes — Produces structured study notes with key concepts, definitions, and takeaways.
• Visualizer — Renders Mermaid diagrams and mind maps to illustrate relationships in the content.
• Deep Explain — Provides a thorough, step-by-step conceptual breakdown of complex topics.
• Chat — Ask follow-up questions about any piece of content in a threaded chat panel.
• Multi-source Input — Accepts URLs (via Firecrawl), YouTube links (via Supadata), uploaded PDFs, or raw text.
• Credits System — Usage-based credits managed per user.
• GitHub OAuth — Sign in with GitHub in addition to email/password.
• Email Notifications — Transactional email via SMTP (auto-configured Ethereal in development).

Tech Stack

Frontend

Layer	Technology
Framework	React 19
Build	Vite 7
Styling	Tailwind CSS 4, Shadcn/UI (Radix Primitives)
Animation	Framer Motion
Diagrams	Mermaid
Charts	Recharts
Drag-and-drop	dnd-kit
Forms	React Hook Form + Zod
Markdown	react-markdown (with GFM + math via KaTeX)
Export	html2canvas, jsPDF
Routing	React Router 7
HTTP	Axios
Icons	Lucide React, Tabler Icons

Backend

Layer	Technology
Runtime	Node.js
Framework	Express 5
Database	PostgreSQL via Supabase
ORM	Prisma 6
Queue	BullMQ + Redis (ioredis)
Text AI	OpenRouter (model-swappable)
Script/TTS AI	Google Gemini (Flash)
TTS	Gemini TTS · ElevenLabs · Deepgram
Web Scraping	Firecrawl
YouTube Transcripts	Supadata
PDF Parsing	pdf-parse
Storage	Supabase Storage
Auth	JWT + bcrypt + GitHub OAuth
Email	Nodemailer (SMTP)
Security	Helmet, express-rate-limit, Zod

Architecture

flowchart TD
    User(["👤 User"])

    subgraph Frontend["Frontend (React + Vite)"]
        SPA["React SPA"]
    end

    subgraph Backend["Backend (Node.js + Express)"]
        API["API Server\n(Express 5)"]
        PQ["Podcast Queue\n(BullMQ)"]
        GQ["Gossip Queue\n(BullMQ)"]
        PW["Podcast Worker"]
        GW["Gossip Worker"]
    end

    subgraph Infrastructure["Infrastructure"]
        Redis[("Redis")]
        PG[("PostgreSQL\n(Supabase)")]
        Storage[("Supabase\nStorage")]
    end

    subgraph AIProviders["AI Providers"]
        OR["OpenRouter\n(text generation)"]
        Gemini["Google Gemini\n(scripts · TTS)"]
        EL["ElevenLabs\n(TTS)"]
        DG["Deepgram\n(TTS)"]
        FC["Firecrawl\n(web scraping)"]
        SD["Supadata\n(YouTube transcripts)"]
    end

    User -- "HTTPS" --> SPA
    SPA -- "REST API" --> API
    API -- "auth / data" --> PG
    API -- "enqueue job" --> PQ & GQ
    PQ --> Redis
    GQ --> Redis
    Redis --> PW & GW
    PW & GW -- "store audio" --> Storage
    PW & GW -- "update job status" --> PG
    API --> OR & Gemini & FC & SD
    PW --> EL & DG & Gemini
    GW --> EL & Gemini

Loading

Long-running tasks (audio generation, script synthesis) are dispatched to background workers via Redis queues so the API responds immediately and the client polls for results.

Getting Started

Prerequisites

• Node.js v20+
• Redis running locally or via a cloud provider
• Accounts / API keys for the external services listed in the environment variables section

Installation

1. Clone the repository

   git clone https://github.com/nyxsky404/Explainer-AI.git
   cd Explainer-AI

2. Install dependencies

   # Backend
   cd backend && npm install
   
   # Frontend
   cd ../frontend && npm install

3. Configure environment variables

   Copy .env.example in the backend/ directory to .env and fill in values:

   # ── Database ──────────────────────────────────────────────────────────────────
   DATABASE_URL="postgresql://USER:PASSWORD@HOST:PORT/DATABASE?schema=public"
   DIRECT_URL="postgresql://USER:PASSWORD@HOST:PORT/DATABASE"
   
   # ── Auth ──────────────────────────────────────────────────────────────────────
   JWT_SECRET="your-very-long-random-secret-here"
   NODE_ENV="development"   # development | production
   
   # ── Server ────────────────────────────────────────────────────────────────────
   PORT=3000
   FRONTEND_URL="http://localhost:5173"
   
   # ── AI Providers ──────────────────────────────────────────────────────────────
   OPENROUTER_API_KEY=""    # https://openrouter.ai/keys
   GEMINI_API_KEY=""        # https://aistudio.google.com/app/apikey
   MODEL="openai/gpt-oss-120b:free"   # OpenRouter model (swap to change globally)
   GEMINI_TEXT_MODEL="gemini-2.5-flash"
   GEMINI_SCRIPT_MODEL="gemini-3-flash-preview"
   GEMINI_TTS_MODEL="gemini-3.1-flash-tts-preview"
   
   # ── OAuth ─────────────────────────────────────────────────────────────────────
   GITHUB_CLIENT_ID=""
   GITHUB_CLIENT_SECRET=""
   
   # ── Storage (Supabase) ────────────────────────────────────────────────────────
   SUPABASE_URL=""
   SUPABASE_SERVICE_KEY=""  # Use the Service Role key, not the anon key
   
   # ── Cache / Queue ─────────────────────────────────────────────────────────────
   REDIS_URL="redis://localhost:6379"
   
   # ── Email / SMTP ──────────────────────────────────────────────────────────────
   # In development, an Ethereal test account is auto-created — no config needed.
   # In production set all of the following:
   SMTP_HOST=""
   SMTP_PORT="587"
   SMTP_SECURE="false"      # true for port 465
   SMTP_USER=""
   SMTP_PASS=""
   SMTP_FROM=""             # e.g. noreply@yourapp.com
   
   # ── Audio TTS (optional alternatives to Gemini TTS) ───────────────────────────
   ELEVENLABS_API_KEY=""    # https://elevenlabs.io
   DEEPGRAM_API_KEY=""      # https://deepgram.com
   
   # ── Web Scraping ───────────────────────────────────────────────────────────────
   FIRECRAWL_API_KEY=""     # https://firecrawl.dev  — required for URL summarization
   SUPADATA_API_KEY=""      # https://supadata.ai    — required for YouTube transcripts

   Create frontend/.env:

   VITE_API_URL=http://localhost:3000/api

4. Initialize the database

   cd backend
   npx prisma migrate deploy
   npx prisma generate

Running the Application

You need three terminals — the API server and both queue workers must run concurrently.

# Terminal 1 — API server
cd backend && npm run dev

# Terminal 2 — Podcast worker
cd backend && npm run worker

# Terminal 3 — Frontend
cd frontend && npm run dev

Open http://localhost:5173 in your browser.

Contributing

Contributions are welcome. Fork the repository and open a pull request for any improvements or bug fixes.