v15

latestOpenAPI 3.1.0Apache-2.0raw.githubusercontent.com2026-08-06121934.4 KB
Text to speech processing

Create speech audio

Generates speech audio from text. Registered only when MODEL_SERVICE=tts. The initial C++ implementation streams audio/wav.

post/v1/audio/speech

Request body

textstring required

Text to synthesize into speech.

descriptionstring

Optional natural-language voice description.

Response

Streaming WAV audio