top of page


- Artificial Intelligent Data Capturing

- Classification Engine that uses Statistical Operators

- Polyglot Accuracy

- Scalable Multi-tasking

- An all-in-one input automation platform

“One of our global shared services-clients in India has more than 10,000 new vendors every year, many of them sending a single invoice. With IRIS HQ OCR and its free-form approach, no templates need to be adjusted and the capturing results are consistently impressive!”

0606CF8F-54C6-4F14-8796-17CD1DE7D804 2.J

Pooja Dargan

Sales & Marketing


IRISXtract for Documents is the flagship of I.R.I.S.’ document understanding technologies. Mainstream solution scenarios empowered by this Intelligent Document Recognition platform are invoice or purchase order capturing and the automated validation of such data, classification of incoming document streams and general data capture from any kind of forms. Enabled by state-of-the-art proprietary artificial intelligent technologies and innovative machine learning engines the software doesn’t require layout specific designs. IRISXtractTM for Documents is intelligent enough to recognize business-critical information autonomously. Such Intelligent Document Recognition is the pre-requisite for robotic process automation since RPA requires fully structured and labeled data.


Artificial Intelligent Data Capturing

The data extraction is based on an artificial intelligent free-form approach that analyses the full-text by considering the context of data. The XContext engine doesn’t require layout and template specific designs, manual user configuration, or complicated layout-based training but allows launching pre-configured solution scenarios. Such solution packages are for example available for invoice processing or purchase order processing and even grant an exceptional user experience right from the start. Our solution experts would appreciate to demonstrate this forward-looking “plug-and-process” approach and discuss with you your individual requirements.


IRISXtract™ for Documents includes an innovative machine learning-based classification engine called XClassify that uses statistical operators, based on certain features and characteristic values, to analyze documents and return with their class. Only an accurate document classification triggers the right process for the respective document class. The XFingerprint technology further utilizes patterns of your documents for classification and is beside the XContext engine an important base technology for automated document separation in case one file that you receive includes several different documents. If you like IRISXtractTM for Documents generates searchable pdfs for your further processing and even compress such pdfs by the I.R.I.S. patented iHQC technology to a minimum.

Polyglot Accuracy

Don’t worry about different languages. IRISXtractTM for Documents utilizes the I.R.I.S. own OCR engine machine written text covering more than 140 languages including Arabic, simplified or traditional Chinese, Japanese, Vietnamese, Thai, and many, many more. The OCR engine “iDRS” itself is avant-garde technology by its HQ-OCR approach which is based on forward-looking multi-layer neural network technology. Accuracy is key! Thus, in the case of handwriting IRISXtractTM for Documents votes between different ICR engines to ensure the best result.

Scalable Multi-tasking

IRISXtract™ for Documents is prepared for both needs of multi-national enterprises and their shared service centers: Multi-Application and Multi-Tenancy. It runs different application scenarios for different purposes (multi-application) and in different configurations for similar purposes (multi-tenancy). IRISXtract™ for Documents is a cloud-ready highly scalable IDR platform with automated performance elasticity that follows custom performance needs whether processing just 10 or 10 000 documents a day. Control your document processing with the help of the I.R.I.S. tracking and tracing component. This statistical concept illustrates by dashboards where your documents are, have been, how long, and reports in detail what has happened during the process whether your need such information just for continuous performance improvement or for audits by your customers or government.

Complete Platform

All this makes IRISXtractTM for Documents your one-stop-shop, your all-in-one input automation platform. You decide if your digital transformation journey is enabling just 10°, 90°,180° or even 360° Content-to-Process! Relevant information is thus, available almost on runtime, resulting in reduced operating costs.

Utilizing IRISXtractTM for Documents leads to increased reliability and process quality, improves the process transparency and auditability, and sets the scene for your business development by unleashing your productivity potentials. With IRISXtract™ for Documents, your document based incoming data is where it ought to be: in the process! With IRISXtractTM for Documents, the business process optimization starts in your digital mailroom.

Digitalization Journey

The technology is one aspect of the Digitalization journey. The application of technology and domain knowledge is equally essential. Even digital data can be optimized at the beginning of the business processes to help systems make informed decisions automatically. Simple digitization projects can be extended to include use cases like Entity Extraction, Sanction Screening, Straight Through Processing, fuzzy decision making, and so on.

Accounts Payable Capture Solution


Free-form approach

I.R.I.S.’ template-free technology reads and captures all types of invoice formats. It allows for an unlimited number of suppliers and adapts automatically when suppliers change their document layouts. Everything across more than 140 script-languages, with excellent accuracy thanks to HQ-OCR-technology.

IRIS’ state-of-the-art Table Finder extracts invoice line items with a precision never seen before, all without any template-requirement.

Rather than relying on static geometric locators (“specific places on the specific vendor’s invoice), IRISXtract™ for Documents invoice processing solution analyses the content after extraction and determines the relevance and meaning; regardless of the vendor’s invoice layout. If invoices come in the form of searchable PDFs (with text layer), IRIS can bypass raw recognition and leverage on the 100%-character accuracy of the text layer to perform the same extraction.

Easy configuration

Due to our many years of working with global shared services centers, our solutions are today deployed for almost all countries worldwide. This implies that we probably already have a ready solution with the relevant business rules for you – wherever you are.


Sales Order Capture Solution

Significantly reduce your workload by automating the sales order processing in your company. IRISXtract™ for Documents captures all relevant order data, delivering streamlined document flows. This solution optimizes your efficiency and customer satisfaction.


IRIS’ Sales Order solution captures relevant customer information from orders and compares it with reference data to complete what’s missing. Identifying the customer is key to trigger the right workflow based on the relevant Service Level Agreement (SLA).

Using the same Freeform technology as the Accounts Payable solution, there is no hassle with different formats and templates.


Any other image- and data-capturing solution can be addressed with IRISXtractTM for Documents and our project-experience. Be it the capturing of forms, Payment Advices, HR-files, legal documents, mail-content etc.


Revisiting our archives and presenting the recording of the special webinar with I.R.I.S. AG talking about Process Automation and Artificial Intelligence. 

bottom of page