Product introduction
Unified OCR provides unified recognition and general document extraction for various types of documents.
Unified OCR integrates recognition capabilities for 59 different scenarios. This feature meets diverse needs and simplifies integration.
General Document Extraction uses a Large Language Model (LLM) to provide general key-value (KV) extraction for long-tail documents.
Product features
Unified OCR
A single API covers six major scenarios: general text recognition, personal ID document recognition, vehicle and logistics document recognition, receipt and credential recognition, business qualification recognition, and mixed document recognition.
General Document Extraction
This feature is a multimodal document understanding application that combines OCR with an LLM. It provides a general extraction service for long-tail documents and cards that standard OCR does not support. Because the service requires no training or configuration, it improves business automation.
Key advantages
High real-time performance
The service runs on Alibaba Cloud's Elastic Algorithm Service (EAS) online clusters and uses continuously optimized inference technology. This provides a low-latency service with elastic scaling.
Stable service
The service scales elastically based on call volume and has high extensibility. Continuous algorithm iterations and optimizations do not affect service stability.
Contact us
Contact us if you have any requirements or feedback, or if you want to speak with our engineers.
You can contact us in the following ways:
Join the official DingTalk group: 35208328
Email business inquiries to ocr_support@list.alibaba-inc.com and include your contact information.