Text-to-Speech
POST
/kling/v1/audio/tts
Generate speech from text using an official Kling voice.
Authorizations
bearer
Type
HTTP (bearer)
Request Body
application/json
Text to synthesize; maximum length 1000 characters. If the text is too long or invalid, the API returns an error code.
Voice ID. The platform provides multiple official voices; see the official documentation for voice effects, IDs, and supported languages. Voice preview does not support custom text. Preview file naming: voice name#voice ID#language. Only official voices are supported; custom voices are not supported.
Language for synthesis (e.g. zh, en).
Speaking rate; valid range 0.8–2.0, rounded to one decimal place
Responses
OK
application/json