A new approach to full text transcription, featuring state-of-the-art recognition engines that allow for printed and cursive handwritten text recognition, without a dictionary

With the single goal to help businesses access more data and deliver more profitable returns from their document conversion and automation processes, TextReader™ features a new approach to full-text transcription and information automation.

Full Text Transcription and Information Automation

For the first time on the market, the same powerful engine can be used for printed and cursive text recognition, enabling all types of documents be transformed into searchable and editable formats – without the use of a dictionary. Powered by a new and unique RNN-based technology developed by Mitek’s in-house R&D Team, users gain complete control over recognition settings and results, and can return both a literal transcription and data extraction from any format of information. Gain added recognition with for specific workflows and data-sets with a custom or trade dictionary and language modeling.

By offering different interfaces and a “plug-and-play” integration for developers, TextReader™ allows for a simple and new approach to the market.

Product Overview

  • Full text recognition technology based on RNN technology – delivering a new approach to the market.
  • Language support for English, French, Arabic, German, Portuguese, Spanish, Chinese (Simplified & Traditional), and Russian
  • Additional languages in development.
  • Cursive recognition without a dictionary; Enhance results further with a user-defined dictionary.
  • Extended character support, alpha numeric capabilities.
  • Many image formats supported, including: JPEG, BMP, TIFF, PDF.

(Click to enlarge)

TextReader™ is the ideal toolkit for organizations looking to automate full text recognition and speed access to data, such as those in the following markets:

  • Banking and Fintech
  • Telecom and Utilities
  • BPO Service Bureaus
  • Healthcare
  • Insurance
  • Government