Document Understanding - Digitization Service Public Preview

Create automations using the framework capabilities of Document Understanding not only via RPA - newly, also via REST APIs hosted on the cloud, up & running in a matter of minutes :dancer:t2:

In this sense, we’re happy to announce the launch of our first service: the Digitization Service - allowing, as the name suggest, to read the text from images or PDFs - offering support for various OCR types, used on demand, defaulting to the in-house built UiPath Document OCR. Having the Digitization Services available as REST API, enables to you access this capability of the DU framework without the requirement of workflows or robots – just by simply calling the corresponding endpoints.

Our initial preview release includes the ability to post a Digitization Request referencing an input file and retrieve the digitized result in the form of a Document Object Model (similar to the current output of the Digitize Activity part of the IntelligentOCR package). In order to gracefully support large files, the service provides the ability to run the operation asynchronously, allowing consumers to ping for the response, available 24 hours for download.

Similar to the activities, the service offers support for various OCR types (with more to come), or if no OCR is required, an option to just turn it off.
Supported OCRs:

  • UiPath Document OCR
  • Google Cloud Vision OCR
  • Microsoft OCR

The service is discoverable via a Swagger interface which can be accessed from the Document Understanding App.
More details about its usage can be found in our official documentation.

Any feedback you may have is really welcome - we’re striving to improve the experience as we go :hammer_and_wrench:


This is an awesome update, congratulations to everyone who worked to pull this new feature through!