Amazon Textract
#199 · Level 3 · Modern architecture · AI and machine learningConcept
Section titled “Concept”Amazon Textract pulls text, tables and form data out of scanned documents. Normal OCR only reads the text, but Textract also understands the structure of the text. For example, in a form, which is the field name and which is its value, or what is in which row and column of a table. You can also ask questions about a document to pull out specific information. It is a fully managed service, so you do not have to build or train any ML model yourself.
Simple example
Section titled “Simple example”A skilled typist who looks at a paper form and enters all the information into the computer, each in the right box.
Why it’s needed
Section titled “Why it’s needed”Typing invoices, forms and receipts by hand is slow and error-prone.
Where it’s used
Section titled “Where it’s used”Processing invoices, reading admission forms, pulling information from bank documents.
Azure equivalent
Section titled “Azure equivalent”Azure AI Document Intelligence
Learn more
Section titled “Learn more”For details and the latest information, see the official documentation: