Optical Character Recognition (OCR) converts static, image-only PDFs — scanned paper documents, for example — into documents with a searchable, selectable text layer.
This is essential for any workflow where scanned documents need to be searched, indexed, or have their text extracted later.
- Endpoint family: /ocr/v1.
- OCR accuracy depends on scan quality; low-resolution or skewed scans will produce more recognition errors.