Audio Generation
Generate speech and dialogue through the unified asynchronous audio API.
- ElevenLabs Multilingual V2 & Turbo V2.5 — multilingual text-to-speech with built-in voices, voice controls, and Unicode character billing
- MiniMax Speech 2.8 — long-text TTS with HD and Turbo models, system or account-owned custom voices, and multiple audio formats
- ElevenLabs Text-to-Dialogue v3 / v4 — multi-speaker MP3 generation at 24 credits per 1,000 Unicode characters
Create tasks with POST /v1/audio/generations, then use GET /v1/tasks/{taskId} or a callback to receive the result.