Choose silence, scene sound or dialogue
Native audio is sound generated with the new video. It is different from uploading an existing recording or adding a soundtrack to a finished clip. SuperSeedance currently accepts text and, in image modes, one picture; it has no audio-upload field, voice picker or dedicated lip-sync editor.
| Goal | Native Audio | Prompt direction |
|---|---|---|
| Completely silent video | Off | Describe the picture and movement. |
| Environment and action sounds | On | Name the audible sources; specify no speech or music if needed. |
| A character speaking | On | Identify the speaker, write a short line and describe delivery. |
| Your own recorded voice or soundtrack | Use a separate editor | Create the visuals here, then assemble the sound externally. |
Set up a short video with audio on
Open Text to Video and choose Seedance 2.5. Set five seconds, 480p and 16:9, turn Native Audio on and leave Draft off.
Review the prompt, audio switch and credit quote before submitting. Reused tasks and links can restore different settings. Opening the template fills the workspace without generating a video or spending credits.
Open the five-second audio templateTemplate 1: give the scene an audible environment
Name the main sound, then a quieter background layer. Tie each sound to something in the scene. Start with a small number of sources so you can hear whether the result follows the request.
One continuous five-second shot of a small fountain in a quiet stone courtyard. Water falls into a shallow basin while a few leaves move in the breeze. Eye-level camera, gentle sideways movement, soft morning light. Audio: close splashing water and faint rustling leaves. No speech or music.Template 2: write one short spoken line
Separate the speaker’s action from the words to be spoken. Use quotation marks for the line and describe its delivery in ordinary language. A short line leaves room for a pause and the visible action; it does not guarantee exact wording or synchronized mouth movement.
One continuous five-second medium shot of a fictional animated lighthouse keeper beside a window. She turns toward the camera and says, calmly, "The storm is over." Soft daylight, steady camera. Audio: her spoken line and faint wind outside. No music or subtitles.Template 3: request an effect without a music track
Describe what causes the sound and when it happens. Keeping audio on allows the effect to be generated. Turning audio off would remove the whole generated soundtrack, including any dialogue you wanted to keep.
One continuous five-second close-up of a wooden toy train rolling across a table and stopping against a small wooden block. Side-on view, locked camera, warm daylight. Audio: soft wheel rattles during movement, one quiet wooden knock at contact, then silence. Only these scene sounds; no voice or background music.Listen to the whole result before making another
Play the complete downloaded video with sound enabled. Check words, pronunciation, timing, unwanted voices, music and the opening or closing sound. Watch for subtitles if you did not request them.
The provider’s guide documents unwanted background music despite instructions excluding it, and notes that subtitles cannot always be prevented. Audio instructions guide generation; they are not separate mixing controls. For another attempt, choose one change and a spending limit. A completed creative miss differs from a failed generation job.
Check task failures and creative missesWhy does my Seedance video have no sound?
Check the player’s mute button, browser or tab mute, system volume and selected speakers. Download the original file and try another player. A muted preview does not reveal whether the file contains an audio track.
Check the original task’s Native Audio setting. Off means a silent output is expected. If it was on and the downloaded file still has no audible sound, keep the task ID and contact support before spending again. If your player displays media information, note whether an audio stream exists: a silent stream differs from no stream.
Changing the workspace switch will not add sound to a completed video. You can keep usable visuals and add sound in an external editor.
What if Seedance adds music anyway?
State the allowed sounds explicitly and remove conflicting soundtrack language from the prompt. For a dialogue shot, distinguish the spoken line and room sound from background music. These adjustments are suggestions to try, not verified fixes for the templates above.
If exact dialogue, a particular recording or clean separate tracks are essential, plan an external audio edit. Removing music from a mixed soundtrack can also damage speech or effects; isolating vocals may discard the effects you wanted. Decide whether to keep the generated track, replace it, or stop before paying for more attempts.
Questions and answers
Does Seedance 2.5 generate audio?+
The model supports generating sound with video. In SuperSeedance, enable Native Audio before submitting; a new workspace starts with it off.
Does “no music” make the video silent?+
No. It asks to exclude music while other sounds may remain. For a silent generation, turn Native Audio off.
Can I upload a recording or choose a voice here?+
No. SuperSeedance currently provides no audio upload or voice picker. An image reference does not lock a voice across clips.
Will the dialogue and lip sync be exact?+
Neither is guaranteed. Listen for changed words and pronunciation, watch mouth timing, and review each result in the language you intend to publish.
Are these prompts tested audio examples?+
No. They are editable starting templates. Review the result and its credit cost before deciding whether another attempt is worthwhile.
Sources
These original prompt templates have not been generated or audio-tested. Controls and provider guidance were checked on 24 October 2026.