Skip to main content

Usage guide

Setup, environment variables, and Python/JavaScript/Go usage examples.

​
AzureVoiceLive

Realtime speech-to-speech integration for Azure Voice Live. Wraps the Azure Voice Live realtime model so it can be used as a full speech-to-speech provider within the runtime. Configuration is supplied through an AzureVoiceLiveConfig instance and resolved into a flat parameter map for the runtime.

​
Constructor

​
api_key
str | None
Azure Voice Live API key. When omitted, the value of the AZURE_VOICE_LIVE_API_KEY environment variable is used.
​
config
AzureVoiceLiveConfig | None
Provider configuration. When omitted, a default AzureVoiceLiveConfig is used.
​
endpoint
str | None
Service endpoint URL. When omitted, the endpoint from config is used, falling back to the AZURE_VOICE_LIVE_ENDPOINT environment variable, then an empty string.

​
aclose

Release any resources held by the integration. This implementation performs no work and exists so the integration can be used interchangeably with providers that require teardown.

​
cleanup

Clean up the integration by delegating to aclose.

​
AzureVoiceLiveConfig

​
Fields

​
model
str
default:"gpt-4o-realtime-preview"
​
voice
str
default:"en-US-AvaNeural"
​
endpoint
Optional[str]
​
modalities
List[str]
default:"…"
​
temperature
Optional[float]
​
max_response_output_tokens
Union[int, str, None]
​
turn_detection
Optional[TurnDetectionConfig]
default:"…"
​
input_audio_transcription
Optional[InputAudioTranscriptionConfig]
default:"…"
​
tool_choice
str
default:"auto"

​
TurnDetectionConfig

​
Fields

​
type
str
default:"server_vad"
​
threshold
float
default:"0.5"
​
prefix_padding_ms
int
default:"300"
​
silence_duration_ms
int
default:"500"
​
create_response
bool
default:"True"
​
interrupt_response
bool
default:"True"

​
InputAudioTranscriptionConfig

​
Fields

​
model
str
default:"gpt-4o-mini-transcribe"