The Transcript Intelligence

The Transcript Intelligence is a Streamlit-powered web application that lets you ask questions about any YouTube video with English captions. By leveraging LangChain, FAISS, HuggingFace embeddings, and Perplexity API, it retrieves answers only from the video transcript, saving you from watching long videos.

---

🚀 Features

• 🔗 Paste a YouTube video URL
• 📑 Fetches and processes the transcript automatically (English captions only)
• ❓ Ask natural language questions about the video
• ⚡ Uses RAG (Retrieval-Augmented Generation) for accurate responses
• 🎨 Modern dark-themed UI with animations

---

🛠️ Tech Stack

• Python
• Streamlit (UI framework)
• YouTube Transcript API (to fetch captions)
• LangChain + FAISS (for semantic search & retrieval)
• HuggingFace Embeddings
• ChatPerplexity (LLM)

---

⚙️ Setup Instructions

1️⃣ Clone the repository

git clone https://github.com/your-username/transcript-intelligence.git
cd transcript-intelligence

2️⃣ Create and activate a virtual environment

python -m venv venv
source venv/bin/activate   # On Mac/Linux
venv\Scripts\activate      # On Windows

3️⃣ Install dependencies

pip install -r requirements.txt

4️⃣ Set up API keys

🔑 Get a Perplexity API Key:

• Go to https://docs.perplexity.ai
• Sign in or create an account
• Generate a new API key from the dashboard

🔑 Get a HuggingFace API Token:

• Go to https://huggingface.co/settings/tokens
• Create a token with read access

5️⃣ Add API Keys

Create a .env file in the project root and add your keys:

# Perplexity API Key
PERPLEXITY_API_KEY=your_perplexity_api_key_here

# HuggingFace API Key
HUGGINGFACEHUB_API_TOKEN=your_huggingface_api_key_here

6️⃣ Run the app

streamlit run Youtube_transcript_project.py

---

📽️ Watch Project Demo

▶️ Watch Demo Video (Google Drive)