Skip to main content
Query parameters are set during the WebSocket upgrade request: 
wss://api.telnyx.com/v2/text-to-speech/speech?voice=Telnyx.NaturalHD.astra

​
Required

​
voice
string
required
Voice identifier in Provider.Model.VoiceId format. Examples: Telnyx.NaturalHD.astra, aws.Polly.Generative.Lucia, azure.en-US-AvaMultilingualNeural. See Engines & Voices.

​
Audio

​
audio_format
string
Output audio format. Most providers return their default format (typically MP3). See Audio Formats for a full breakdown of default formats and supported options per provider.

​
Caching

​
disable_cache
boolean
Set disable_cache to true to bypass the server-side audio cache and force fresh synthesis for every request. When false (the default), identical text + voice + settings combinations return a cached result, reducing latency and cost for repeated phrases.

​
Voice Settings

Provider-specific parameters (speed, pitch, emotion, etc.) are passed in the voice_settings object of the handshake frame, not as query parameters. See Voice Settings.