Azure AI Services

When to invoke

• "Add vector or hybrid search to my app with Azure AI Search."
• "Transcribe this audio or convert text to speech."
• "Extract fields from these documents with Document Intelligence."
• "Which Azure AI service should I use for semantic search?"

Services

Service	Use When	MCP Tools	CLI
AI Search	Full-text, vector, hybrid search	azure__search	az search
Speech	Speech-to-text, text-to-speech	azure__speech	-
OpenAI	GPT models, embeddings, DALL-E	-	az cognitiveservices
Document Intelligence	Form extraction, OCR	-	-

MCP Server (Preferred)

When Azure MCP is enabled:

AI Search

• azure__search with command search_index_list - List search indexes
• azure__search with command search_index_get - Get index details
• azure__search with command search_query - Query search index

Speech

• azure__speech with command speech_transcribe - Speech to text
• azure__speech with command speech_synthesize - Text to speech

If Azure MCP is not enabled: Run /azure:setup or enable via /mcp.

AI Search Capabilities

Feature	Description
Full-text search	Linguistic analysis, stemming
Vector search	Semantic similarity with embeddings
Hybrid search	Combined keyword + vector
AI enrichment	Entity extraction, OCR, sentiment

Speech Capabilities

Feature	Description
Speech-to-text	Real-time and batch transcription
Text-to-speech	Neural voices, SSML support
Speaker diarization	Identify who spoke when
Custom models	Domain-specific vocabulary

SDK Quick References

For programmatic access to these services, see the condensed SDK guides:

• AI Search: Python | TypeScript | .NET
• OpenAI: .NET
• Vision: Python | Java
• Transcription: Python
• Translation: Python | TypeScript
• Document Intelligence: .NET | TypeScript
• Content Safety: Python | TypeScript | Java

Service Details

For deep documentation on specific services:

• AI Search indexing and queries -> Azure AI Search documentation
• Speech transcription patterns -> Azure AI Speech documentation

Output template

## Azure AI service result

**Status:** designed | implemented | blocked
**Summary:** <one sentence covering scope and outcome>

### Details
Chosen service, index or model configuration, and calling code changes.

### Validation
- <check performed>: <result and evidence>

Quality gate

• The selected service matches the retrieval, speech, or extraction need.
• Keys are replaced by managed identity or Entra auth wherever supported.
• The output follows ## Output template exactly.
• Every reported check was performed and its evidence is shown.
• Irreversible Azure actions were confirmed with the user first.