Azure

Azure Speech Services provides 400+ neural voices across 140+ languages, making it the widest multilingual TTS offering available. It supports SSML for granular speech control and offers enterprise-grade reliability with global availability. The go-to choice for multilingual applications requiring broad language and regional accent coverage.
Usage
To use Azure as the TTS engine, pass the following JSON in the TTSConfig field of the StartAIConversation API:
// json — TTSConfig
{
  "TTSType": "azure",
  "SubscriptionKey": "<your_azure_subscription_key>",
  "Region": "southeastasia",
  "VoiceName": "en-US-AmandaMultilingualNeural",
  "Language": "en-US",
  "Rate": 1
}
For the complete TTSConfig parameter reference, see the Text-to-Speech Configuration.
Parameter reference
Field
Type
Required
Description
TTSType
String
Yes
Must be "azure".
SubscriptionKey
String
Yes
Azure Speech resource subscription key. Obtain from Azure Portal.
Region
String
Yes
Subscription region (e.g., southeastasia, eastus). See Azure TTS regions.
VoiceName
String
Yes
Voice name. See Azure language and voice support.
Language
String
Yes
Language for TTS (e.g., en-US, zh-CN).
Rate
Float
No
Speech speed. Range: [0.5–2]. Default: 1.
For more details on Azure TTS, see the Azure Text-to-Speech documentation.
Next step: StartAIConversation API Reference﻿
﻿

Field	Type	Required	Description
`TTSType`	String	Yes	Must be `"azure"`.
`SubscriptionKey`	String	Yes	Azure Speech resource subscription key. Obtain from Azure Portal.
`Region`	String	Yes	Subscription region (e.g., `southeastasia`, `eastus`). See Azure TTS regions.
`VoiceName`	String	Yes	Voice name. See Azure language and voice support.
`Language`	String	Yes	Language for TTS (e.g., `en-US`, `zh-CN`).
`Rate`	Float	No	Speech speed. Range: [0.5–2]. Default: 1.