Azure

Azure is a text-to-speech plugin. It synthesizes the LLM’s text response into natural-sounding speech using Azure Neural TTS voices.

Setup

Set your Azure Speech resource key in the worker environment. Create a Speech resource and key in the Azure portal:

export AZURE_SPEECH_KEY=<key>

Import the plugin and pass it to the pipeline’s tts slot.

from zrt.plugins import AzureTTS

tts = AzureTTS()

# Pipeline(tts=tts, ...)

Parameter	Type	Default	Description
`speech_key`	`Optional[str]`	`None`	Azure Speech resource subscription key. Falls back to the `AZURE_SPEECH_KEY` environment variable when unset.
`speech_region`	`Optional[str]`	`None`	Azure region identifier for the Speech resource (e.g. `"eastus"`, `"westeurope"`, `"southeastasia"`). Falls back to the `AZURE_REGION` environment variable, then `"eastus"` if that is also unset. Must match the region where your Azure Speech resource was created, since keys are region-scoped.
`voice`	`str`	`"en-US-JennyNeural"`	Azure Neural TTS voice name in `"<locale>-<Name>Neural"` format. Common options include `"en-US-AriaNeural"`, `"en-US-GuyNeural"`, `"en-GB-SoniaNeural"`, `"fr-FR-DeniseNeural"`, `"de-DE-KatjaNeural"`, `"ja-JP-NanamiNeural"`.
`sample_rate`	`int`	`24000`	PCM output sample rate in Hz. Azure Neural TTS natively supports `24000` (standard) and `48000` (high-fidelity).

The synthesized audio is streamed back to the caller as the LLM produces its response.