Tasks and inputs
StepAudio 3 Gen Preview is listed for Audio generation, Text-to-speech, Music Generation. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Oct 8, 2026
StepAudio 3 Gen Preview creates a complete audio scene from text, audio references and role descriptions.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
StepAudio 3 Gen Preview creates a complete audio scene from text, audio references and role descriptions.
StepAudio 3 Gen can combine character voices, sound effects, ambience, background music and singing in one audio result. The brief can describe roles, delivery and when different elements should occur. This makes it a candidate for a scene draft, a game dialogue sequence or a short narrated production with coordinated sound.
Specify who speaks, the intended line, the emotion and the order of sounds. Separate an effect that should occur during dialogue from one that should follow it. For several roles, describe each voice consistently throughout the prompt. If the result misses a line or an important cue, revise that requirement explicitly instead of simply requesting better audio.
The published endpoint is POST /v1/audio/generate, using stepaudio-3-gen-preview. Inputs include text and audio references; the output is audio. The documentation describes temporary trial access. Permanent billing figures remain unknown here, and a production client should verify access and result handling against the audio-generation API.
Listen for omitted lines, speaker confusion, audible transitions and whether background sound obscures speech. Compare the result with the written timeline, including the closing moment. Keep the brief and references with the chosen clip so another revision can reproduce the intended roles and structure.
Compare the model on the work you need to complete.
StepAudio 3 Gen Preview is listed for Audio generation, Text-to-speech, Music Generation. It accepts text, audio and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
Compare pronunciation, voice consistency and time to the first playable audio. Check the service’s billing unit for the actual script.
Use the model identifier and account requirements for the selected API route. Confirm its request limits and output format with one complete task before connecting a larger workflow.