Text-to-speech
POST /tts/speech
POST
/tts/speech
Synthesizes speech from text and streams audio. Follows the OpenAI audio/speech API convention. Returns chunked audio stream.
Request Body required
Section titled “Request Body required ”object
model
required
TTS model (currently only “kokoro”)
string
input
required
Text to synthesize
string
voice
required
Voice name or blend expression
string
response_format
string
speed
number format: double
Responses
Section titled “ Responses ”Streaming audio
string format: binary
Invalid request
object
error
required
string
message
required
string
details
object
key
additional properties
any