VidQA is an AutoGPT app that summarises and answers questions about arbitrary YouTube videos using LangChain and LLMs.
๐๏ธ Under construction. More coming soon...
๐ก The main idea for this app came from a YouTube video by Dave Ebbelaar.
- Set up your virtual env with all the required dependencies
poetry install
- Set your environment variables in a file called
.env
(see.env.example
) - Create a resource and deploy a model using Azure OpenAI
Note: You can use this directly with OpenAI's API (without using Azure's OpenAI Services) with minor modifications to the code and environment variables
TODO: Add Streamlit app link
Note: Whe using notebooks, make sure you activate the correct poetry environment. You can set your Python kernel to the value that gets returned when you run
poetry run which python
.
- Browse the example notebooks in
notebooks/