Soniox

Voice format: Soniox.tts-rt-v2.<VoiceId> Soniox is a third-party TTS provider with real-time, streamed synthesis. Every Soniox voice can speak any of the supported languages, and each voice has its own accent, for example British, Brazilian Portuguese, or Japanese: choose the voice for its sound and accent, and set language to the language of the text.

Voices

Soniox offers a catalog of around 200 voices under tts-rt-v2. Browse the full catalog via the Voices API or the available voices page:

GET https://api.telnyx.com/v2/text-to-speech/voices?provider=soniox

Samples of five of them:

Voice	Gender	Voice ID	Description
Adrian	Male	`Soniox.tts-rt-v2.Adrian`	Deep, composed, professional. Neutral accent.
Daniel	Male	`Soniox.tts-rt-v2.Daniel`	Low, steady British baritone. Suits long narration.
Emma	Female	`Soniox.tts-rt-v2.Emma`	Casual British delivery. Conversational.
Grace	Female	`Soniox.tts-rt-v2.Grace`	Unhurried, assured American voice. Suits lessons and support scripts.
Nina	Female	`Soniox.tts-rt-v2.Nina`	Young, upbeat British voice. Suits casual conversation.

Models

Model	Description
`tts-rt-v2`	Real-time model. Default and the only model accepted.

Language

Set language to the two-letter ISO 639-1 code of the text, for example en, it, or pt. The default is en. SSML is not supported. Send plain text.

Voice Settings

Field	Type	Default	Description
`language`	string	`en`	Two-letter ISO 639-1 code of the text.
`speed`	float	`1.0`	Speaking rate, `0.7` to `1.3`.
`reduce_silence`	boolean	`false`	Shortens the pauses between words.
`audio_format`	string	`mp3`	`mp3`, `wav`, `pcm_s16le`, `pcm_mulaw`, or `pcm_alaw`.
`sample_rate`	integer	format default	`8000`, `16000`, `24000`, `44100`, or `48000`. See Audio Formats & Sample Rates.

Limits

A request produces up to two minutes of audio. Split longer text into several requests.

WebSocket

Soniox is available on the TTS WebSocket API. Select the voice with the voice query parameter and pick the output with audio_format and sample_rate. Send language, speed, and reduce_silence in voice_settings on the init frame.

wss://api.telnyx.com/v2/text-to-speech/speech?voice=Soniox.tts-rt-v2.Emma&audio_format=mulaw

{
  "text": " ",
  "voice_settings": {
    "language": "it",
    "speed": 1.0,
    "reduce_silence": false
  }
}

See the Soniox WebSocket provider page for the accepted formats and sample rates.

REST API

Fields

Field	Type	Default	Description
`language`	string	`en`	Two-letter ISO 639-1 code of the text.
`speed`	float	`1.0`	Speaking rate, `0.7` to `1.3`.
`reduce_silence`	boolean	`false`	Shortens the pauses between words.
`audio_format`	string	`mp3`	`mp3`, `wav`, `pcm_s16le`, `pcm_mulaw`, or `pcm_alaw`.
`sample_rate`	integer	format default	`8000`, `16000`, `24000`, `44100`, or `48000`.
`output_type`	string	`binary_output`	`binary_output`, `base64_output`, or `audio_id`.

{
  "text": "Buongiorno, come posso aiutarla?",
  "voice": "Soniox.tts-rt-v2.Emma",
  "voice_settings": {
    "language": "it",
    "speed": 1.0,
    "audio_format": "mp3",
    "sample_rate": 24000
  }
}

Response

Default (binary_output): chunked audio bytes. With output_type: "base64_output": JSON with base64-encoded audio. With output_type: "audio_id": JSON with an audio_url for deferred retrieval.

Audio Formats & Sample Rates

Format	Accepted Sample Rates (Hz)	Default
`mp3`	16000, 24000, 44100, 48000	24000
`wav`, `pcm_s16le`	8000, 16000, 24000, 44100, 48000	24000
`pcm_mulaw`, `pcm_alaw`	8000	8000

In-Call Playback

Soniox voices work with the Call Control speak, gather_using_speak, and conference speak commands, and with TeXML <Say>.

Send plain text: payload_type must be text.
Set language to the two-letter ISO 639-1 code of the text, for example it.
voice_settings accepts speed and reduce_silence.

{
  "payload": "Buongiorno, come posso aiutarla?",
  "voice": "Soniox.tts-rt-v2.Emma",
  "language": "it",
  "voice_settings": {
    "speed": 1.0
  }
}

<Response>
  <Say voice="Soniox.tts-rt-v2.Emma" language="it">Buongiorno, come posso aiutarla?</Say>
</Response>

See In-Call Playback for details.

AI Assistants

Choose a Soniox voice such as Soniox.tts-rt-v2.Emma in the assistant’s voice settings. voice_speed maps to the Soniox speed setting and accepts 0.7 to 1.3. Soniox voices speak every supported language, so one voice can serve an assistant that changes language during a call.

AI Assistants

Build voice AI assistants using Soniox voices.

TTS REST API

Generate speech directly with REST TTS requests.

WebSocket Streaming

REST API

Providers

Other

API Reference

For AI Agents

Voices

Models

Language

Voice Settings

Limits

WebSocket

REST API

Fields

Response

Audio Formats & Sample Rates

In-Call Playback

AI Assistants

AI Assistants

TTS REST API

​Voices

​Models

​Language

​Voice Settings

​Limits

​WebSocket

​REST API

​Fields

​Response

​Audio Formats & Sample Rates

​In-Call Playback

​AI Assistants

AI Assistants

TTS REST API

Voices

Models

Language

Voice Settings

Limits

WebSocket

REST API

Fields

Response

Audio Formats & Sample Rates

In-Call Playback

AI Assistants