StepAudio 3 Gen
StepAudio 3 Gen creates audio from role descriptions, scripts and natural-language instructions. A request can combine several speakers with effects, background sound, music and singing.
Preview
Describe the order of sounds
The model supports instructions for when sound elements should appear and in what order. Role descriptions can specify voice, style, emotion and interaction.StepFun stepaudio-3-gen guide
Write a scene rather than a voice label
Practical guidance. State who speaks, what happens around them and when each sound starts. Use TTS when the task needs only one spoken track. A combined audio scene gives less separation for later mixing than independently produced tracks.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Preview |
| Access | API, Provider product |
| Model identifiers | stepaudio-3-gen-previewstepaudio-3-gen |
| Current access | The preview name is documented now. The paid stepaudio-3-gen name is scheduled for 9 November 2026.StepFun billing catalog and 9 November 2026 audio change |
| Endpoint | POST /v1/audio/generate.StepFun stepaudio-3-gen guide |
Related models
- StepAudio 3 Music
StepFun · StepAudio 3
StepAudio 3 Music supports full songs, instrumental music and accompaniment for an isolated vocal.
Launch date not confirmedPreviewAPI · Provider product - StepAudio 3 TTS
StepFun · StepAudio 3
StepAudio 3 TTS produces speech from text and reference audio.
Launch date not confirmedAPI · Provider product