Skip to main content

NLP Solutions

This domain covers building applications that understand, analyze, and generate human language using Azure AI Language and Speech services. It represents 15–20% of the AI-102 exam.

You'll implement text analytics (sentiment, entities, key phrases), PII detection and redaction, language translation, speech-to-text and text-to-speech, and conversational AI with CLU (Conversational Language Understanding) and Custom Question Answering. These services form the backbone of chatbots, virtual assistants, and content processing pipelines.

The exam heavily tests SDK patterns for these services. Know how to initialize clients, send batch requests, handle async operations, and parse structured results. Also know the rebranding: LUIS → CLU, QnA Maker → Custom Question Answering.

What You'll Learn​

  • Analyze text for sentiment, entities, key phrases, and language
  • Detect and redact personally identifiable information (PII)
  • Translate text and documents across 100+ languages
  • Convert speech to text with real-time and batch transcription
  • Generate natural speech from text with neural voices
  • Build conversational AI with intents and entities (CLU)
  • Create knowledge bases with Custom Question Answering
  • Implement real-time speech translation

Skills Measured​

  • Analyze text using Azure AI Language (sentiment, NER, key phrases)
  • Detect and redact PII from text
  • Translate text and documents with Azure AI Translator
  • Implement speech-to-text and text-to-speech
  • Build and deploy Conversational Language Understanding (CLU) models
  • Create and manage Custom Question Answering projects
  • Implement speech translation

Challenges​

#TitleKey Topics
31Text Analytics: Sentiment & EntitiesTextAnalyticsClient, batch analysis, confidence scores
32Key Phrases & Language DetectionExtraction, multi-language support, opinion mining
33PII Detection & RedactionPII categories, redacted text, domain-specific filters
34Text & Document TranslationTranslator API, custom dictionaries, document translation
35Speech-to-TextReal-time recognition, batch transcription, custom models
36Text-to-SpeechNeural voices, SSML, custom neural voice
37Conversational Language Understanding (CLU)Intents, entities, training, deployment, testing
38Custom Question AnsweringKnowledge bases, multi-turn, active learning, chit-chat
39Speech TranslationReal-time multi-language translation, audio output

Prerequisites​

  • Completed Domain 1 (Plan & Manage) or equivalent knowledge
  • Azure AI Services multi-service resource provisioned
  • Python or C# programming skills
  • Basic understanding of NLP concepts (helpful but not required)