SSE

Stream text-to-speech audio as JSONL (JSON lines) objects over HTTP. A less performant alternative to WebSockets, without text input streaming.

Headers

X-API-KeystringRequired

Request

This endpoint expects an object.
transcriptstringRequired
Text for narration.
voiceobjectRequired
Voice for narration.
output_formatobjectOptional
Audio format specification.

Response

This endpoint returns a stream of object.
chunkobject
OR
errorobject