Posted On: Apr 3, 2023

Amazon Textract is a managed machine learning service that automatically extracts text, handwriting, and data from scanned documents. We regularly improve the accuracy of the underlying machine learning models and add new features based on customer feedback. Today, we are pleased to announce new features and accuracy enhancements for the AnalyzeDocument - Tables feature to help customers automate their document processing workflows. 

The new tables model update adds support for four new elements commonly found within tabular structures in documents: table title, section titles, table footers and summary cells. Customers will be able to directly extract these elements from tabular structures and save time writing and maintaining complex post-processing code. Additionally, it also provides a way to identify the type of table. Lastly, we are pleased to announce overall accuracy improvements on documents commonly found in the insurance, financial services, mortgages and manufacturing industries. 

This update will be available in US East (Ohio, N. Virginia), US West (N. California), US West (Oregon), Asia Pacific (Mumbai, Seoul, Singapore, Sydney), Canada (Central), Europe (Frankfurt, Ireland, London, Paris), and the AWS GovCloud (US-East, US-West) Regions.

To get started, log on to the Amazon Textract console to try out the new features. To learn more about Textract capabilities, please visit the Amazon Textract website, developer guide, or resources page.