====== Cognitive Services Overview ======

Azure Cognitive Services is a collection of pre-built APIs and SDKs that enable developers to add AI capabilities to their applications—without requiring deep knowledge of machine learning or data science.

These services are divided into categories that mimic human cognitive abilities such as vision, language, speech, and decision-making.

===== 🧠 Core Categories of Cognitive Services =====

**1. Vision**  
  * Image analysis, object detection, facial recognition.  
  * Example: An app that detects emotions from facial expressions.

**2. Speech**  
  * Speech-to-text, text-to-speech, translation, speaker recognition.  
  * Example: A multilingual voice assistant using speech translation.

**3. Language**  
  * Natural language understanding, text analytics, QnA Maker.  
  * Example: A chatbot that extracts meaning from user questions.

**4. Decision**  
  * Anomaly detection, content moderation, personalizer.  
  * Example: Detecting fraudulent transactions in real time.

**5. Search**  
  * Bing APIs for web search, image search, news search, etc.  
  * Example: Integrating custom web search into an app.

===== 🔑 Key Features =====

  * **No ML expertise required** – Pretrained models are accessible via REST APIs.
  * **Scalable and secure** – Hosted in Azure, with built-in compliance tools.
  * **Can be combined** – You can integrate multiple services (e.g., vision + language).

===== 🧪 In the AI-900 Exam =====

You should be able to:

  * Identify scenarios where Cognitive Services are a good fit.
  * Differentiate between the service categories.
  * Recognize when to use a pre-built solution vs. custom ML models.

===== 📚 Related Pages =====

  * [[AI:AI900:azure_services:machine_learning|Machine Learning in Azure]]
  * [[AI:AI900:tools:azure_ml_studio|Azure ML Studio]]