youtube extender

Inspiration

Have you ever felt regret after wasting time on a YouTube video, only to find out that it offered nothing new or interesting? To solve this, I decided to use an LLM to provide a summary of the video, so you can get an idea of what it contains before deciding whether it's worth watching, rather than being lured in by clickbait. Additionally, for educational videos, we offer the option to generate a quiz based on the content, so you can test your understanding and ensure it wasn’t a waste of time.

What it does

Currently, you can summarize videos, create quizzes based on the content, and convert videos into articles. For content transformation, we require more processing power for video processing, so we only provide the timestamp.

How we built it

You can test it at https://talk-to-video.streamlit.app/. To use the app, you'll need an API key, which you can get here https://cloud.sambanova.ai/apis . The app was built with Streamlit for the user interface.

Challenges we ran into

We wanted to add a video Q&A feature. It worked locally, but there was an issue when deploying it to Streamlit Cloud, so we had to remove it from the code.

What's next for youtube extender

We are working on building something similar to NotebookLM, combining documents, videos, and articles. We're also aiming to create a proper content repurposer, where you can directly generate short-form videos from longer ones. Currently, you can only view the timestamp and content details.

Built With

• sambanova
• streamlit