https://github.com/aadi2084/youtube-assistant
A Streamlit-powered AI assistant that allows users to ask questions about any YouTube video based on its transcript. It leverages LangChain, FAISS, and OpenAI embeddings to generate relevant responses from video content.
https://github.com/aadi2084/youtube-assistant
ai-assistant faiss langchain llm machine-learning mistral-7b nlp openai openrouter python sentence-transformers streamlit vector-database youtube
Last synced: 3 months ago
JSON representation
A Streamlit-powered AI assistant that allows users to ask questions about any YouTube video based on its transcript. It leverages LangChain, FAISS, and OpenAI embeddings to generate relevant responses from video content.
- Host: GitHub
- URL: https://github.com/aadi2084/youtube-assistant
- Owner: Aadi2084
- License: mit
- Created: 2025-03-05T11:37:44.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-03-05T12:15:03.000Z (over 1 year ago)
- Last Synced: 2025-03-05T12:30:08.232Z (over 1 year ago)
- Topics: ai-assistant, faiss, langchain, llm, machine-learning, mistral-7b, nlp, openai, openrouter, python, sentence-transformers, streamlit, vector-database, youtube
- Language: Python
- Homepage:
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🎥 youtube-assistant
YouTube Assistant is a Streamlit-powered web app that allows users to ask questions about YouTube videos based on their transcripts. It utilizes **LangChain**, **FAISS**, and **Sentence Transformers** to retrieve and process relevant video content.
## 🚀 Getting Started
Clone the repository:
`git clone https://github.com/your-username/youtube-assistant.git && cd youtube-assistant`
Install dependencies:
`pip install -r requirements.txt`
Set up API keys in a `.env` file:
OPENAI_API_KEY=your_openai_api_key
OPENROUTER_API_KEY=your_openrouter_api_key
Run the app:
`streamlit run main.py`
## 📌 Deployment on Streamlit Cloud
```
Push your code to GitHub:
git add .
git commit -m "Initial commit"
git push origin main
```
Then, deploy via [Streamlit Community Cloud](https://share.streamlit.io/) by linking your GitHub repository.
## 🛠Technologies Used
- **Python** (Backend)
- **Streamlit** (UI Framework)
- **LangChain** (LLM Integration)
- **FAISS** (Vector Database)
- **Sentence Transformers** (Embeddings)
- **Mistral-7B via OpenRouter API** (LLM)
- **YouTubeLoader** (Extracts video transcripts)