Ashutosh Mishra 发布 AI-VIDEO-ASSISSTANT：基于 RAG 的多语言视频问答助手

An AI-powered Video Assistant that automatically downloads YouTube videos or processes local video/audio files, transcribes speech, generates summaries, extracts action items, and enables multilingual Question Answering using Retrieval-Augmented Generation (RAG).

🚀 Features

• 🎥 Process YouTube videos
• 📂 Process local audio/video files
• 🎙️ Speech-to-Text using Whisper & Sarvam AI
• 📝 Automatic Transcript Generation
• 📄 AI-Powered Summary
• 📌 AI-Generated Title
• ✅ Extract Action Items
• 🔑 Extract Key Decisions
• ❓ Extract Open Questions
• 🤖 RAG-based Question Answering
• 🌍 Multilingual Support
  • English
  • Hindi
  • Hinglish

🛠 Tech Stack

• Python
• LangChain
• Mistral AI
• OpenAI Whisper
• Sarvam AI
• ChromaDB
• HuggingFace Embeddings
• yt-dlp
• FFmpeg

📂 Project Structure

AI-VIDEO-ASSISSTANT
│
├── core/
│   ├── extractor.py
│   ├── rag_engine.py
│   ├── summarizer.py
│   ├── transcriber.py
│   └── vector_store.py
│
├── utils/
│   └── audio_processor.py
│
├── main.py
├── app.py
├── Requirements.txt
└── README.md

⚙️ Installation

Clone the repository

git clone https://github.com/ashutoshmishra7275mishra-create/AI-VIDEO-ASSISSTANT.git

cd AI-VIDEO-ASSISSTANT

Install dependencies

pip install -r Requirements.txt

Install FFmpeg

winget install "FFmpeg (Essentials Build)"

Create a .env file

SARVAM_API_KEY=YOUR_SARVAM_API_KEY
MISTRAL_API_KEY=YOUR_MISTRAL_API_KEY

▶️ Run

python main.py

Example

Enter YouTube URL or local file path:
https://www.youtube.com/watch?v=jNQXAC9IVRw

💬 Example Questions

English

• What is this video about?
• Summarize the video.
• What are the key points?
• What did the speaker say?

Hindi

• वीडियो किस बारे में है?
• सारांश बताइए।
• मुख्य बिंदु क्या हैं?

Hinglish

• Video kis baare me hai?
• Summary batao.
• Speaker ne kya bola?

📸 Output

• Transcript
• AI Summary
• AI Title
• Action Items
• Key Decisions
• Open Questions
• Multilingual AI Chat

🎯 Future Improvements

• Streamlit Web UI
• PDF Export
• Timestamp-based Answers
• Speaker Diarization
• Chat History
• Semantic Search
• Multiple Video Support

👨‍💻 Author

Ashutosh Mishra

GitHub: https://github.com/ashutoshmishra7275mishra-create

⭐ Support

If you like this project, don't forget to ⭐ Star the repository.