Explanation:
Box 1: Azure Document Intelligence in Foundry Tools
Extracting structured-data forms and invoices.
Azure Document Intelligence (formerly known as Azure Form Recognizer) within the Foundry Tools ecosystem can be used to extract structured data, including key-value pairs, tables, and specific fields from forms and invoices. It is designed to transform unstructured or semi-structured data from PDFs, images, and other files into actionable, structured JSON output.
Box 2: Azure Language in Foundry Tools
Summarizing written content from business reports.
Azure Language in Foundry Tools (formerly Azure AI Language) includes specific features for summarizing written content, including business reports, in both extractive and abstractive formats.
Key Summarization Capabilities
Native Document Summarization: This feature can directly parse and summarize files in their original formats, such as PDF, Word (DOCX), and plain text.
Summarization Approaches:
-Extractive: Selects the most important original sentences from the document to create a summary.
-Abstractive: Generates new, concise sentences that capture the main idea without directly copying the source text.
Powered by Advanced Models: The service utilizes Large Language Models (LLMs) and Small Language Models (SLMs), such as GPT-4o and Phi-3.5-mini, to provide high-quality, low-latency results.
Box 3: Azure Vision in Foundry Tools
Generate descriptive text for uploaded images.
Azure Vision in Foundry Tools (part of Azure AI Services within the Foundry ecosystem) can be used to analyze uploaded images and automatically generate descriptive, human-readable text.
This capability is part of the Image Analysis feature, which generates English-language captions describing the content of an image.
Key aspects of this functionality include:
*-> Descriptive Captions: The service generates complete sentences based on objects identified in the image, providing multiple options ordered by a confidence score.
*-> Image Tagging: It can generate a list of words (tags) identifying objects, beings, scenery, or actions.
Reference:
https://azure.microsoft.com/en-us/products/ai-foundry/tools/document-intelligence
https://azure.microsoft.com/en-us/products/ai-foundry/tools/vision
