Speech API Reference

input

true

file

The text to generate audio for. The maximum length is 4096 characters.

model / deployment-id

true

string

Model ID to use for the request. The available TTS models are gpt-4o-mini-tts.

Voice

true

string

The voice to use when generating the audio. Supported voices are alloy, echo, fable, onyx, nova, shimmer, coral, verse, ballad, ash, sage, marin, cedar, amuch, aster, brook, clover, dan, elan, marilyn, meadow, jazz, rio, megan-wetherall, jade-hardy, megan-wetherall-2025-03-07, and jade-hardy-2025-03-07.

instructions

false

string

Control the voice of your generated audio with additional instructions.

prompt

false

string

Optional text to guide the model's style or continue a previous audio segment. Ensure the prompt matches the audio language.

response_format

false

string

Format of the audio output. Supported output formats: mp3, opus, aac, flac, wav, and pcm. Default output is mp3.

Speed

false

number

The speed of the generated audio. Select a value from 0.25 to 4.0. Default is 1.0.

stream_format

false

string

The format to stream the audio in. Supported formats are sse and audio. Default stream value is audio.