Generate speech audio

Authorization

BearerAuth

AuthorizationBearer <token>

In: header

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

input*string

The text to synthesize (max 1000 chars for gpro/gpro31/xai, 500 for orpheus)

Length1 <= length <= 1000

model?string

The voice model to use when selecting a voice by name. Omit when using voiceId.

Value in

"gpro"
"gpro31"
"orpheus"
"xai"

response_format?string

Audio format. Default depends on model

Value in

"wav"
"mp3"

seed?integer

Optional deterministic seed for providers that support it (e.g. Gemini)

Range-9007199254740991 <= value <= 9007199254740991

speed?number

Speech speed multiplier for Grok (xai) voices. Range 0.7-1.5. Ignored by other models.

Range0.7 <= value <= 1.5

style?string

Emotion/style variant (e.g., "happy", "sad", "whisper")

temperature?number

Sampling temperature for Gemini voices (gpro/gpro31). Range 0-2; higher is more expressive. Ignored by other models.

Range0 <= value <= 2

voice?string

Voice name. Use with model, or pass voiceId instead. See GET /api/v1/voices for available voices.

Length1 <= length

voiceId?string

Voice ID from GET /api/v1/voices. Use instead of voice + model.

Length1 <= length

Response Body

`application/json`

curl -X POST "https://example.com/api/v1/speech" \  -H "Content-Type: application/json" \  -d '{    "input": "Hello, world!",    "model": "gpro",    "voice": "achernar"  }'

{  "credits_remaining": 0,  "credits_used": 0,  "url": "http://example.com",  "usage": {    "input_characters": -9007199254740991,    "model": "string"  }}

Generate speech audio

Authorization

Request Body

Response Body

200application/json

400application/json

401application/json

402application/json

404application/json

422application/json

429application/json

500application/json

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`