A legal firm has thousands of scanned paper documents stored as image files in Amazon S3. The firm needs to extract printed text, handwriting, and structured data from these documents to make them searchable and enable automated processing of forms and tables. Which AWS service should they use?