Complete Guide to AI Services on Azure: From Applied AI to ML Platforms
In today’s digital world, artificial intelligence (AI) plays an increasingly central role in almost every aspect of our lives. Whether you’re a developer looking to integrate AI functionality into your application, a business owner looking to optimise operations, or even a curious subject matter expert wanting to know more about what AI can offer, it’s essential to understand the different services available to implement these technologies. Microsoft Azure offers a range of AI services that can meet a variety of needs and skill levels. This guide aims to break down these offerings into understandable categories, from applied AI, to cognitive services, to the ML platform, to help you navigate the ever-evolving AI ecosystem on Azure.
I. Azure AI Applied Services:
These are pre-built AI services designed to solve specific business problems without requiring deep expertise in machine learning. They aim to reduce the cost and time associated with training and maintaining models.
- Azure Form Recognizer: Recognizes and extracts information from forms and documents.
- Azure Cognitive Search: AI-enhanced search service for smarter application development.
- Azure Video Indexer: Analyzes, extracts metadata, and indexes videos.
- Azure Immersive Reader: Makes textual content more accessible and engaging.
- Azure Bot Service: Creates intelligent bots to interact with users across multiple channels.
- Azure Metrics Advisor: Performance monitoring and analytics service for anomaly detection.
II. Cognitive Services:
Cognitive services offer a range of AI capabilities designed to deliver human-like intelligence to applications. These services enable your applications to see, hear, speak, understand and even make decisions. They come with pre-built models that can be easily integrated into applications.
1. OpenAI Service Preview: Applies advanced coding and language models to a variety of use cases.
2. Language:
- Entity Recognition: Identifies commonly-used and domain-specific terms.
- Sentiment Analysis: Automatically detects sentiments and opinions from text.
- Question Answering: Transforms information into easy-to-navigate questions and answers.
- Conversational Language Understanding: Enables your applications to interact with users in natural language.
- Translator: Translates more than 100 languages and dialects.
3. Vision:
- Computer Vision: Analyzes the content in images and videos.
- Custom Vision: Customizes image recognition to meet your business needs.
- Face API: Detects and identifies people in images.
4. Speech:
- Speech to Text: Transcribes audible speech into readable, searchable text.
- Text to Speech: Converts text to lifelike speech for more natural interfaces.
- Speech Translation: Integrates real-time speech translation into your apps.
5. Decision:
- Anomaly Detector: Identifies potential problems early on.
- Content Moderator: Detects potentially offensive or unwanted content.
- Personalizer: Creates rich, personalized experiences for every user.
III. Azure ML Platform
The Azure ML platform is designed to cover the entire machine learning lifecycle. It offers solutions for all skill levels, from minimal to intensive code, while supporting open source frameworks and languages. It can also manage deployment, explainability and MLOps.
1. Low Code:
- Automated ML: For people who need machine learning models but are not machine learning experts. It automates the model creation process.
- Drag-and-drop designer: Allows you to create machine learning models using a drag-and-drop graphical interface.
2. Code Centric:
- Azure Notebooks: For developers who prefer to code their own models, Azure offers notebooks to facilitate development.
- Azure Databricks: A data platform based on Apache Spark, ideal for large-scale machine learning tasks.
- Azure Data Science Virtual Machines (DSVM): VMs pre-configured with data science and machine learning tools.
3. No Code:
- Azure Automated ML UI: A user interface that lets you create machine learning models without writing a single line of code.
- Azure Machine Learning Studio: A visual tool for data science that requires no coding skills.
Azure AI Services — https://azure.microsoft.com/en-us/solutions/ai
