https://github.com/benitomartin/youtube-llm

LLM Q&A and Summarization App
https://github.com/benitomartin/youtube-llm

chromadb langchain python streamlit whisper

Last synced: 7 days ago
JSON representation

LLM Q&A and Summarization App

• Host: GitHub
• URL: https://github.com/benitomartin/youtube-llm
• Owner: benitomartin
• Created: 2023-10-12T10:10:48.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-04-20T18:08:40.000Z (9 months ago)
• Last Synced: 2024-11-08T10:10:01.525Z (2 months ago)
• Topics: chromadb, langchain, python, streamlit, whisper
• Language: Python
• Homepage:
• Size: 658 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# YOUTUBE Q&A AND SUMMARIZATION APP 📺



    



This repository hosts an app developed using **Whisper** and **Langchain** that allows the creation of a Q&A assistant and video summarization. The model's maximum context length is 4097 tokens (gpt-3.5-turbo).

The App can be run locally but requires an `OPENAI_API_KEY` in the `.env` file. Feel free to ⭐ and clone this repo 😉

## 👨‍💻 **Tech Stack**

![Visual Studio Code](https://img.shields.io/badge/Visual%20Studio%20Code-0078d7.svg?style=for-the-badge&logo=visual-studio-code&logoColor=white)

![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)

![OpenAI](https://img.shields.io/badge/OpenAI-74aa9c?style=for-the-badge&logo=openai&logoColor=white)

![Linux](https://img.shields.io/badge/Linux-FCC624?style=for-the-badge&logo=linux&logoColor=black)

![Git](https://img.shields.io/badge/git-%23F05033.svg?style=for-the-badge&logo=git&logoColor=white)

![Streamlit](https://img.shields.io/badge/Streamlit-FF4B4B?style=for-the-badge&logo=Streamlit&logoColor=white)

## 💬 Set Up

 I recommend installing the modules in the following order. The `ffmpeg` module is required for the proper functioning of the application. You can install it using Conda as follows:

```bash

conda install -c conda-forge ffmpeg

```

```bash

pip install git+https://github.com/openai/whisper.git

```

```bash

pip install -r requirements.txt

```

## 🫵 App Deployment

The app can be used running `streamlit run app.py` in the terminal. There are 2 options on the sidebar, Q&A or Summarize. I recommend using videos no longer than 5 min of speech due to the model tokens' limitations.

The first option allows a Q&A assistant to ask questions about the video.



    



The second option allows us to get a summary of the video.