Usage guide
Setup, environment variables, and Python/JavaScript/Go usage examples.
AzureVoiceLive
Speech-to-speech conversation with the Azure Voice Live API, handling listening, reasoning and speaking in one realtime model. Choose it for low-latency voice agents on Azure. Authenticates with AZURE_VOICE_LIVE_API_KEY from your environment, or an explicitapi_key.
Fields
str | None
default:"UNSET"
Your Azure Voice Live resource endpoint URL. Falls back to the AZURE_VOICE_LIVE_ENDPOINT environment variable. Defaults to None (not set).
str
required
Realtime model to converse with, e.g. “gpt-4o-realtime-preview”. Required.
Mapping[str, Any] | None
default:"UNSET"
Session settings such as voice, modalities, turn_detection and temperature, as a mapping. Defaults to None (not set).