AI-based classification:
Advanced classification algorithms leverage modern Machine Learning and Natural Language Processing technologies and offer highest document classification quality together with more flexible classification options, new classification modes and improved classification API

New input formats: Office documents
Office document formats can be now processed in addition to the image input formats and PDF input formats - Windows version only.

Extraction of data from Machine Readeble Zones (MRZ) in ID documents - Windows version only.

Improved accuracy of Japanese OCR:
Improved accuracy together with the new “Japanese Modern” OCR language. New 'special predefined language' for enhanced recognition of dates, times, addresses and names

Faster recognition of Chinese & Korean
Due to the usage of the newly trained Convolusional Neural Network (CNN) for Asian OCR languages - Windows version only.

New deployment in the Cloud:
New type of license 'Online license' supports deployment within the Cloud environment (e.g. services like Amazon EC2 and Microsoft Azure), virtual environments and Docker containers.

Improved layout reconstruction:
Improved tables reconstruction, detection and recreation of balanced text columns, improved layout retention on TXT export.

New export formats: HTML 5, ALTO 3.1 (the latest ALTO XML scheme is supported)

New PDF saving options:
The latest PDF 2.0 standard support & export to PDF in accordance with PDF/UA standard support. In addition, in the Windows version a broader set of tags for tagged PDF export formats is available.

New PDF/A saving options:
PDF/A-2b and PDF/A-3b support

New XML saving options and improvements: \\Faster export to XML, direct export of list elements and export an information about tab-space characters.

Read more: FineReader Engine 12 - What is new Overview