Case-studies
Recent Case-studies
These are some of the case studies that were achieved by Canspirit team.
Base OCR
creating SOP Documentation And Charts From Video
This project focuses on automating SOP documentation from training videos using AI. By leveraging Google’s Gemini AI, NLP, and computer vision, it extracts key information from videos, converts it into structured documents, and generates workflow diagrams. This solution enhances efficiency, ensures accuracy, and reduces manual effort, making documentation faster and more reliable for businesses.
- NLP-BERT – For entity extraction and text processing.
- TensorFlow 2.0 & Keras – For AI-powered data analysis and structuring.
- Computer Vision – For image and video recognition
- Python Libraries – Pillow, Tesseract, and EasyOCR for text extraction.
- Flask Framework – For deployment and automation.
Finance
Invoice Data Extraction to cSV using AI
The project automates the extraction of invoice data using OCR and LLMs, transforming invoices into structured CSV files for easy integration with CRM and ERP systems. It removes the need for manual data entry, cutting processing time by 70% and enhancing accuracy and scalability. Developed with Python, AI models, and OCR tools like Tesseract, the system provides efficient and error-free financial automation for businesses.
- Tesseract
- AI Model
- Python
Base OCR
Text based detection from Images
You can use these sections to highlight the features of heading. Use these paragraphs to focus on the topic you want. Make sure you keep it short and attractive.
- Python {Natural Language Processing – BERT entity extraction method}.
- Ubiai tool to augment data
- Tensorflow-2.0, Keras-api to extract textual data
- Python libraries – Pillow to read images, tesseract, EasyOCR
- Poppler – To convert pdf to image
- Python Framework – Flask {For deployment}
Finance
Predictive Analytics for Bank Loan Default
Loan Distribution was a principal offering of the Bank. The major earnings of the Bank come from Loans disbursed and the interest earned. The Bank offered personal and company loans. The Bank wanted to reduce the credit risk and the defaults in loan repayment.
- R
- R Studio
- dplyr for data preprocessing
Finance
Predictive Analytics for Bank Loan Default
Loan Distribution was a principal offering of the Bank. The major earnings of the Bank come from Loans disbursed and the interest earned. The Bank offered personal and company loans. The Bank wanted to reduce the credit risk and the defaults in loan repayment.
- Python
- Pandas, Numpy
- Pandas for data preprocessing