Overview
The Splitter transforms mixed, unordered, or high-volume page batches into fully separated, logically organized documents. Whether pages arrive from scanner feeders as massive PDFs/TIFFs or through APIs and watched folders via the Receiver node, the Splitter uses advanced rules, AI models, and business context to divide documents with precision. It adapts to your structure, your workflows, and your expected document sets.
Capabilities
-
Rule- or AI-Based Splitting
Split batches using content understanding, learned patterns, or deterministic indicators such as separator sheets, specialized markers, or colored pages generated by the Tagger.
-
Landmark Detection
Use specialized markers, separator sheets, or colored pages generated by the Tagger to guide splits.
-
Content-Aware Logic
Analyze text and layout to determine logical document boundaries—ideal for mixed business paperwork.
-
Page Reordering
Automatically resort pages when scanning or upload order is incorrect.
-
AI Coach Training
Teach the Splitter how you want documents separated using examples—no ML expertise required.
-
Expected Document Mode
Split using checklists or pre-defined bundles sourced from APIs, databases, repositories, or targeted web-crawls via the Receiver.
-
Continuous Stream Support
Handles documents arriving in real-time from folders, APIs, mobile uploads, and scanning devices.
Benefits
-
Zero Manual Sorting
Eliminate tedious page-by-page review and ensure every batch is cleanly separated.
-
Accurate Downstream Automation
Perfectly formed documents maximize classification, OCR, extraction, and workflow success.
-
Flexible to Your Workflow
Works with separators, content, metadata, or learned behavior—whichever matches your operations.
-
Consistent Output
Guarantees that every batch yields predictable, standardized document packages for your business systems.
FAQ
Yes. It is optimized for high-volume feeder scans and can process thousands of pages efficiently.
Not necessarily. You can rely on content analysis, AI Coach training, or expected document lists if you prefer a marker-free workflow.
Yes. You can create multiple Splitter configurations tailored to each process or document type.
You simply show the system examples of how your documents should be split. It learns your rules and applies them automatically.
Yes. Originals are always retained and versioned. Reordering only applies to output packages, ensuring auditability.