API Reference

Synthesize speech from text. Body: text (required), voice (required), language, format. Returns audio bytes.

Streaming text-to-speech — same parameters as /v1/audio/speech; audio is returned as it's generated.

Transcribe audio to text. Multipart: audio file (required), language, timestamps (bool). Returns transcript JSON.

Create an asynchronous transcription or synthesis job for longer audio or text. Returns a job ID.

Retrieve the status and result of an async job by its ID.

List voices available to your account. Returns an array of { id, name, language, description }.

List supported models and language codes.

Retrieve account usage — minutes used, minutes included, and overage for the current billing period.