ABBYY has released a new application programming interface (API) named Document AI API, designed to assist developers in converting unstructured business documents into structured and accurate data.
The Document AI API offers a developer-focused interface with minimal setup required and pre-trained models. This can facilitate the creation of proof-of-concepts and the integration of optical character recognition (OCR) technology and intelligent document processing (IDP) into business workflows.
According to ABBYY, the API was specifically created to address the growing need for reliable and consistent data extraction from various business documents. The company states that the new API allows for effortless document transformation using a small amount of code, streamlining the process for developers looking to incorporate document AI into their applications.
Nick Hyatt, Vice President, Engineering R&D at ABBYY, commented, "As a vanguard of OCR, ABBYY has long had a vibrant community of cutting-edge developers creating transformational solutions with our advanced document AI. We are providing them a new API with minimal setup, access to ample community resources, and pre-trained models for building proof-of-concepts. ABBYY Document AI API is a major step forward for developing automated document workflows."
Industry analysis from IDC suggests the intelligent document processing (IDP) market is expanding rapidly. IDC projects the market will grow from USD $2.4 billion in 2023 to USD $10.5 billion in 2028, representing a compound annual growth rate of 34.9%. This growth is attributed to increased cloud adoption, advancements in artificial intelligence, and the expansion of document AI use cases.
Amy Machado, Senior Research Manager, Enterprise Content and Knowledge Management Strategies at IDC, said, "In the age of AI, OCR is experiencing a true renaissance. Developers struggle with extracting reliable data from documents and will often begin with general large language models for this process. However, they quickly face challenges with hallucinations, data inconsistencies, and errors in document processing, and often lack support for multiple languages, handwriting recognition and complex document structures. There is a need for purpose-built solutions specifically designed for document processing that prioritises easy integration, flexibility, scalability, accuracy, and consistency."
The ABBYY Document AI API was initially released as a technical preview. Through the API, developers can use pre-trained models to extract information to accelerate automation for business processes such as Know Your Customer (KYC) procedures, account openings, customs clearance, invoice processing, expense management and order processing.
The solution offers high-precision OCR capabilities that maintain the logical structure of documents, a feature ABBYY argues is essential for producing data suitable for generative AI applications, retrieval-augmented generation (RAG), and sophisticated language model training.
SDKs for commonly used programming languages, including Python, C#, JavaScript, and Java, are available to support integration with the API. ABBYY also emphasises community support, offering resources and a Discord community for developers interested in using Document AI API.
