chapter seven

7 Azure AI

 

This chapter covers

  • Understanding natural language processing, speech, vision, and content moderation using AI
  • Creating a project in Azure AI Foundry to manage the entire AI life cycle of a product
  • Considerations for responsible AI use

Azure AI offers a suite of advanced tools designed to enhance business operations through AI. These services include natural language processing (NLP), speech recognition, and Computer Vision—each empowering organizations to automate processes, improve customer interactions, and gain deeper insights from their data. For example, a customer service chatbot built with Azure’s NLP capabilities can understand and respond to queries in natural language, delivering instant, accurate answers that elevate the overall support experience. Likewise, Azure’s speech-to-text services can transcribe call recordings for quality assurance and training, enabling a retail company to analyze customer feedback and quickly identify common issues.

On the visual side, Azure’s Computer Vision service can automatically analyze and tag images, turning a pile of book-cover photos into searchable metadata. For Banning Books, that means extracting title text from covers, spotting visual motifs such as “circuit board” or “neon cyberpunk,” and finding visually similar titles so customers can browse by style as well as by topic—no more office arguments about whether a cover is “retro” or just “mildly nostalgic”—the model calls it (politely).

7.1 Natural language processing and OpenAI

7.2 Azure Speech services

7.2.1 Speech to Text

7.2.2 Text to Speech

7.2.3 Speech Translation

7.3 Azure Vision services

7.3.1 Computer Vision

7.3.2 Custom Vision

7.3.3 Face API

7.3.4 Content Moderator

7.4 Azure AI Foundry

7.4.1 Starting with AI Foundry

7.4.2 Configuring a data source

7.4.3 Selecting a model

7.4.4 Model fine-tuning

7.4.5 Deployment pipelines and operational governance

7.5 Responsible AI

7.5.1 Fairness

7.5.2 Transparency