Spaces:
Sleeping
Sleeping
Angel
commited on
Commit
·
35be24f
1
Parent(s):
3cbd301
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,11 +1,26 @@
|
|
| 1 |
-
Agentic AI using YouTube Transcript-API & RAG
|
| 2 |
-
Overview
|
| 3 |
-
This project focuses on creating an agentic AI chatbot that leverages YouTube videos as its knowledge base. By utilizing the YouTube Transcript API and Retrieval-Augmented Generation (RAG), the AI bot can extract information from YouTube video transcripts and answer queries based on that content.
|
| 4 |
|
| 5 |
-
|
| 6 |
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Agentic AI using YouTube Transcript-API & RAG
|
|
|
|
|
|
|
| 2 |
|
| 3 |
+
## Overview
|
| 4 |
|
| 5 |
+
This project focuses on creating an **agentic AI chatbot** that leverages **YouTube videos** as its knowledge base. By utilizing the **YouTube Transcript API** and **Retrieval-Augmented Generation (RAG)**, the AI bot can extract information from YouTube video transcripts and answer queries based on that content.
|
| 6 |
+
|
| 7 |
+
The process of fetching new YouTube videos and extracting their transcripts is automated using **GitHub Actions**, ensuring that the knowledge base is continuously updated with fresh content from YouTube.
|
| 8 |
+
|
| 9 |
+
## Features
|
| 10 |
+
|
| 11 |
+
- **Automated Video Fetching**: GitHub Actions automatically fetches new YouTube videos and updates the dataset regularly.
|
| 12 |
+
- **Transcript Extraction**: The **YouTube Transcript API** extracts transcripts from the fetched YouTube videos.
|
| 13 |
+
- **Retrieval-Augmented Generation (RAG)**: The bot uses RAG to query the AI and retrieve information from video transcripts to answer user queries.
|
| 14 |
+
- **Bot Interaction**: A chatbot interface answers questions based on the YouTube video transcripts.
|
| 15 |
+
|
| 16 |
+
## Project Structure
|
| 17 |
+
|
| 18 |
+
```bash
|
| 19 |
+
/data
|
| 20 |
+
/scripts
|
| 21 |
+
- fetch_new_videos.py # Script to check for new YouTube videos
|
| 22 |
+
- fetch_transcripts.py # Script to fetch transcripts for new videos
|
| 23 |
+
/transcripts
|
| 24 |
+
- transcripts.csv # CSV or JSON file where transcripts of videos are stored
|
| 25 |
+
/github-actions
|
| 26 |
+
- fetch_and_update.yml # GitHub Action for fetching new videos and updating transcripts
|