Skip to main content
POST
cURL

Continue building

Write instructions

Choose intent, emotion, and delivery.

Realtime text to speech

Manage instructions across conversation turns.

Convert text to speech

Generate audio without word timing.

Authorizations

xi-api-key
string
header
required

Breeze Developer API key.

Path Parameters

voice_id
string
required

Voice ID to use for speech generation. See List voices.

Body

application/json
output_format
enum<string>
default:pcm

Output audio encoding: pcm or mp3. Default: pcm. See Output formats.

Available options:
pcm,
mp3
model_id
string | null

Model ID for speech generation. When omitted, the server selects a model based on the account rollout and requested language. See List models.

Required string length: 1 - 120
language_code
string | null

ISO 639-1 two-letter language code supported by the selected model. See supported language codes.

Required string length: 2
Pattern: ^[A-Za-z]{2}$
instructions
string | null

Initial performance instructions, written in the same language as the input text. Leading and trailing whitespace is removed. See Updating instructions.

voice_settings
RealtimeTtsVoiceSettingsPayload · object | null

Optional realtime voice settings. See Expressive controls.

inactivity_timeout_seconds
integer | null

Idle timeout for the WebSocket session. Defaults to 30 seconds and is capped at 180 seconds.

Required range: 1 <= x <= 180
enable_logging
boolean | null

Whether realtime turns should be stored in generation history.

Response

Realtime WebSocket session token and compatible transport URLs.

client_secret
string
required

Short-lived token for browser WebSocket connections.

websocket_url
string
required

WebSocket URL including the client_secret query parameter.

expires_at
string
required

ISO-8601 expiration timestamp for the client secret.

direct_websocket_url
string | null

Optional direct WebSocket URL without credentials in its query. Authenticate with the documented WebSocket subprotocols.

audio_format
RealtimeTtsAudioFormat · object

Selected realtime audio stream format.