OCR (Optical Character Recognition)
Also known as: receipt scanning · AI receipt reader · document AI
Definition
OCR (Optical Character Recognition) is AI technology that reads text from images — screenshots, photos of receipts, scanned documents — and converts it into structured, searchable data. In bookkeeping, OCR extracts transaction details from payment screenshots and paper receipts.
Detailed Explanation
OCR works by identifying text regions in an image, recognizing individual characters, and reconstructing words and numbers. Modern AI-powered OCR is significantly more accurate than traditional template-based OCR because it understands context — it knows that '$850.00' next to 'Acme Corporation' on a PayPal screenshot represents a payment amount and sender, not random text. Key OCR capabilities for freelancers: payment platform screenshot recognition (PayPal, Venmo, Stripe, Alipay, WeChat Pay), paper receipt digitization (photograph a crumpled receipt, extract the data), multi-language support (50+ languages), and multi-transaction detection (multiple payments in one screenshot detected individually). OCR accuracy depends on image quality — clear, well-lit screenshots achieve 90%+ accuracy. Paper receipts in poor lighting may be 80-85%.
Freelancer Example
A freelancer receives a payment via PayPal and takes a screenshot of the confirmation. They upload it to Tally Assistant. The AI OCR reads 'Acme Corporation sent you $850.00' and extracts: merchant = Acme Corp, amount = $850.00, date = June 25, 2026, type = Income. The original screenshot is saved as supporting documentation.