Skip to main content
POST
cURL

Continue building

Design a voice

Write a voice description and preview script.

Stream a design preview

Stream one PCM preview.

Save voice preview

Save a reusable voice for TTS.

Write instructions

Choose intent, emotion, and delivery.

Authorizations

xi-api-key
string
header
required

Breeze Developer API key.

Query Parameters

output_format
string | null
default:mp3

Audio format for the embedded preview payloads. Supported values: mp3, wav, flac, pcm, aac, opus. Default: mp3. See Output formats.

Body

application/json
voice_description
string
required

Text description of the desired voice. See Voice Design prompting.

Required string length: 3 - 500
text
string | null

Optional preview script for the generated voice previews. See Audio tags.

Required string length: 3 - 500
model_id
string | null

Model ID for speech generation. Selected automatically when omitted. 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}$
guidance_scale
number | null

Generation guidance strength. Accepted range: 1.0 to 10.0. Random within this range when omitted. See Voice Design controls.

Required range: 1 <= x <= 10
preview_count
integer | null

Number of preview variants to generate. Defaults to 1 and accepts values from 1 to 3.

Required range: 1 <= x <= 3

Response

Generated voice previews.

previews
VoiceDesignPreviewResponse · object[]
required
text
string
required