Introducing the Robotic Document Separator by Haystac

AI that transforms a large file into multiple documents. Without separator sheets.

BOSTON, MA – August 3, 2021

Haystac, a leader in AI and content analytics software, today announced a breakthrough in document processing automation.  The Haystac Robotic Document Separator (RDS) completely replaces the need for separator sheets, and the human activity associated with them. 

Powered by the most advanced AI algorithms and data science, RDS can read any large scanned file, automatically identify where each document begins and ends, separate those documents into individual files, and send that information on to the next stage of the workflow.  Unlike human workers, RDS works around the clock, never tires, never stops for a break, and does not require a benefits package.

Under the Hood

RDS uses AI models that were pre-trained on tens of thousands of diverse documents.  Deep learning / neural network methods were used to train the models.  This allows RDS to capture document experience and to accumulate knowledge, just like humans do.  The pre-trained models use transfer learning so they can work with your documents.

This is a no-code solution, and users don’t have to know a thing about AI or data science to operate RDS.  Haystac’s R&D team has hidden the AI “magic” beneath the simple user interface.

Business Benefits
  • Save money, by scanning without separator sheets.  Costs of labour and printing can range from 6 to 10 cents per document. That adds up very quickly, even for small scan centers. 
  • Faster time to income, because there is no document separation prep.  RDS also trains on far fewer samples than older machine learning or rules-based methods.
  • Consistency of document separation, because the machine is consistent while humans vary.
  • Fully automated software that runs 24×7 without any breaks.

Deployment

  • As a cloud-based micro-service with REST API.
  • Can be integrated into any workflow.
  • Option to install on-prem.
Comments are closed.