How to use this RAG AI Teaching assistant on your own data

Step 1 - Collect your videos

Make a videos folder and Move all your video files to the that folder

Step 2 - Convert to mp3

Convert all the video files to mp3 by ruunning video_to_mp3 and store it in audios folder

Step 3 - Convert mp3 to json

Convert all the mp3 files to json by ruunning mp3_to_json and store it in jsons folder

Step 4 - Convert the json files to Vectors

Use the file preprocess_json to convert the json files to a dataframe with Embeddings and save it as a joblib pickle

Step 5 - Prompt generation and feeding to LLM

Read the joblib file and load it into the memory. Then create a relevant prompt as per the user query and feed it to the LLM