This project is an AI-powered tool to help users efficiently transcribe, summarize, and query long-form audio and video content. The user can upload an audio file or input a link for YouTube. The tool will retrieve or generate the transcript and put it into a Chroma database. A Large Language Model uses Retrieval-Augmented Generation to gather context from the transcript to respond to user prompts.
Whether watching recorded lectures, analyzing interviews, or reviewing online content, this tool streamlines the extraction of key information without the need for sitting through hours of material. Additionally, it can act as a supplement to this material, acting as an interactive assistant that is knowledgeable about the content.
Save time, skipping lengthy playback
Improve productivity, easily searching and retrieving information from videos without rewatching
Enhance learning, especially if you are a college student reviewing a lecture
Interact with the video, asking the LLM-powered chatbot specific questions about its content
Video Transcription: Either pulls a transcript directly from YouTube or can transcribe a user-uploaded audio file using Whisper, running locally on the user’s computer
RAG: the chatbot has access to the information from the video ,allowing it to generate insightful, context-aware responses
Clone the GitHub repo (linked at the bottom)
Install the required dependencies
Make sure you have your API key set up in an environment
Run python streamlit run lecture_summarizer.py
Full code can be found at: https://github.com/d3vmeh/lecture-summarizer
There are no models linked
There are no models linked
There are no datasets linked
There are no datasets linked