Revolutionizing Document Processing GPTOCR API by Blackcoffer
Client Background
Client: A leading tech firm in the USA
Industry Type: IT & Consulting
Products & Services: IT Solutions, Software Development
Organization Size: 100+
The Problem
Design and develop an API as a service backend, the API should be integrated with GPT and OCR technologies to extract documents it should be hosted on Azure
Our Solution
/token – It takes username and password as a input and generate API_key/token to run the other APIs
/api/template/create-template – This is a Post request. It stores the created json template in the database and generates a token id.
/api/document/upload – This api takes a file as an input. We can upload .pdf, .docx, .png, .jpg, .jpeg, .txt files. It has basically 2 parts. We can just upload the document or we can also provide template id to process the uploaded document according to the template id.
/api/document/process – This api takes template id and document id as an input. It fetches the template and document from the database and uses the ocr method to extract the text from the document. This extracted text and template are then processed by gpt api which generates the final output.
/api/template/all – This api fetches all the templates created by the user using create-template api.
/api/template/update-template – This api can update the created template.
/api/template/delete – This api deletes the created template by giving template id.
/api/document/all – This api shows all documents uploaded by user
/api/document/delete – This api deletes the document by document id.
Deliverables
All the APIs on the Azure server
Tools used
fastapi, gpt api, pytessaract, pypdf2
Language/techniques used
fastapi, gpt api, pytessaract, pypdf2, python
Skills used
python, Rest API development
Databases used
MS Sql
Web Cloud Servers used
Azure
What are the technical Challenges Faced during Project Execution
Main challenge in this project extracting text from images and pdfs and generate json output according to template
How the Technical Challenges were Solved
In the apis we can upload .pdf, .docx, .png, .jpg, .jpeg, .txt files. It has basically 2 parts. We can just upload the document or we can also provide template id to process the uploaded document according to the template id.
It fetches the template and document from the database and uses the ocr method to extract the text from the document. This extracted text and template are then processed by gpt api which generates the final output..
Business Impact
This will help users to directly upload any pdf or image and extract useful information in json format.
Project Snapshots
Summarize
Summarized: https://blackcoffer.com/
This project was done by the Blackcoffer Team, a Global IT Consulting firm.
Contact Details
This solution was designed and developed by Blackcoffer Team
Here are my contact details:
Firm Name: Blackcoffer Pvt. Ltd.
Firm Website: www.blackcoffer.com
Firm Address: 4/2, E-Extension, Shaym Vihar Phase 1, New Delhi 110043
Email: [email protected]
Skype: asbidyarthy
WhatsApp: +91 9717367468
Telegram: @asbidyarthy