azure-ai-voicelive-py - Build real-time voice AI applications with Azure SDK
Provides patterns and tools for building real-time voice AI applications using bidirectional WebSocket communication with Azure AI Voice Live SDK.
Tags
Updated: 2026-09-21Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Connect to Azure Voice Live API
- Configure session and audio settings
- Stream base64 PCM16 audio input
- Receive real-time audio and transcripts
- Handle function calling and conversation
Inputs
- Azure Cognitive Services endpoint URL
- Azure credential or API key
- Base64 encoded PCM16 audio data
Outputs
- Real-time audio stream output
- Audio transcription text events
- Function call request events
Requirements
- Python azure-ai-voicelive package
- Python aiohttp package
- Python azure-identity package
- AZURE_COGNITIVE_SERVICES_ENDPOINT environment variable
- Azure gpt-4o-realtime-preview model access