Denver, Colorado, April 15, 2026 (GLOBE NEWSWIRE) -- Apryse, a leader in document processing technology, today announced its Spring 2026 release, including Intelligent Character Recognition (ICR), an AI-powered capability designed to convert handwritten content into structured, machine-readable data. The release targets what many organizations consider to be the final barrier to end-to-end document automation.
While Optical Character Recognition (OCR) has long enabled the digitization of printed text, handwriting has remained largely inaccessible to automation due to its variability and lack of structure. As a result, industries like healthcare, financial services, insurance, and legal continue to rely heavily on manual data entry, slowing down workflows and introducing risk.
Apryse’s ICR SDK technology closes this gap.
“AI is only as powerful as the data it can access, and for most organizations, that data is still locked inside documents” said Andrew Varley, Chief Product Officer at Apryse. “By bringing AI-driven handwriting recognition into our extraction toolkit, we’re enabling developers to work with complete datasets, not partial ones, and build automation that reflects how information actually exists in the real world within modern document applications.”
Unlike API-based extraction tools that require data to be processed externally, Apryse’s ICR capability is built into its SDK architecture, allowing organizations to process sensitive information entirely within their own infrastructure. This is particularly critical for regulated industries where data privacy, security, and compliance are non-negotiable.
The introduction of ICR is part of a broader push by Apryse to help developers transform documents from static files into dynamic, AI-ready data assets and automation workflows across the entire document lifecycle.
Beyond Extraction: Closing the Gaps in Document Workflows
In addition to ICR, the Spring 2026 release introduces several capabilities designed to eliminate friction across document processing and collaboration workflows:
- Email-to-PDF conversion (EML/MSG): Transform email threads into standardized, audit-ready PDFs while preserving metadata and attachments for compliance and long-term record keeping.
- PDF sanitization API: Programmatically remove hidden metadata, embedded scripts, and sensitive information to reduce security risks before documents are shared or archived.
- Expanded file conversion support (including RTF): Bring legacy document formats into modern, searchable PDF workflows with high fidelity across Apryse’s document SDK ecosystem.
From Documents to Data Pipelines
For developers, the implications are significant. Healthcare providers can digitize handwritten intake forms, financial institutions can automate loan and check processing, and insurance teams can streamline claims documentation without manual intervention within their existing software systems.
As enterprises continue to invest in AI and automation workflows, the ability to reliably access and operationalize all forms of document data, including handwriting, is increasingly essential.
By bridging the gap between unstructured inputs and structured outputs, Apryse is positioning documents as a core component of modern data pipelines, not a bottleneck. For the past two years, Apryse has released new or improved extraction features every quarter to ensure the availability of a complete and accurate toolkit for developers.
About Apryse
Apryse is a leading provider of document technology, helping organizations get more value from their documents. Its toolkit supports the full document lifecycle, from high-fidelity viewing and editing to conversion, digital signatures, and intelligent data extraction.
Trusted by more than 20,000 companies, including 85% of the Fortune 100, Apryse powers mission-critical workflows where performance, security, and accuracy matter most.
For more information, visit www.apryse.com.
Attachments
- Apryse Spring Release 2026
- Apryse Introduces Intelligent Character Recognition (ICR) in Spring 2026 Release
