A RAG App using Azure AI Search for interacting with pdf
Project Name
Chat with your pdf!
Description
I made a simple app "Chat with your pdf!" that can interact with a bulky pdf and collect and display a precise answer using all the relevant points throughout the pdf. It is built using the python language and run by streamlit
I made use of the following azure functionalities:
- ☀️ Azure AI Search
- ☀️ Azure Open AI
- ☀️ Azure Blob storage
Here, I have uploaded the pdf from my local device, for a more general use case I can make use of ☀ Azure Form Recognizer Service Step 1: The user asks a question or a query in the app user interface Step 2: The server sends this query to azure AI Search Step 3: The AI Search collects top meaningful results from blob storage which has split and indexed the pdf Step 4: AI Search sends these results along with the query to the server which takes it to Azure Open AI Step 5: Azure Open AI uses this knowledge from the pdf to give a well tailored specific answer to the question rather than a vague one Step 6: The app server get this answers and displays it to the user interface
Technology & Languages
- [ ] JavaScript
- [ ] Java
- [ ] .NET
- [X] Python
- [X] AI Studio
- [X] AI Search
- [ ] PostgreSQL
- [ ] Cosmos DB
- [ ] Azure SQL
Project Repository URL
https://github.com/JoJo1188-hub/RAG-Hackathon.git
Deployed Endpoint URL
No response
Project Video
https://youtu.be/uXxWHWaHCSQ?si=VEUV3N1qaL5TKRRs
Team Members
Josephine Crystal R Mathew
Hello @JoJo1188-hub, thank you for participating in RAG Hack!
The team is working hard to distribute badges. Please have each team member fill out this form: aka.ms/raghack/badge-dist
Thank you!