Skip to main content
POST
Text-to-Speech (SSE)

Authorizations

Authorization
string
header
required

Pass your API key as a Bearer token: Authorization: Bearer sk_996_....

Body

application/json

Same as TTSRequest, but output_format.container must be raw (streaming endpoints cannot emit a WAV header).

model_id
enum<string>
required

The TTS model. Currently gepard-1.0.

Available options:
gepard-1.0
transcript
string
required

The text to synthesize. Billed at 1 credit per character. A single request must stay under the generator's ~46.5s audio ceiling — roughly ~900 characters of en-US prose, but far fewer for number-heavy text (~130 digits), since digits are spoken as whole words. Over the limit → input_too_long (400); split longer content across requests.

voice
object
required

Selects the voice to synthesize. Required — there is no default voice; a request that doesn't resolve to a voice returns an error (it is no longer silently substituted). The only accepted form is the nested object { "mode": "id", "id": "<voice-uuid>" }; a top-level voice_id is not supported.

output_format
object
required

Output audio format. Each field is validated against a fixed set; anything outside these values returns 400 with no silent fallback.

language
string

Optional BCP-47 language tag (full dialect, e.g. en-US).

Example:

"en-US"

Response

A text/event-stream of chunk events followed by a done event.

The response is of type string.