Audio Pad
Add leading and trailing silence to an audio clip, pad it up to a minimum length, and resample it. The defaults add no silence and set no minimum, so out of the box it only resamples to 48 kHz stereo mp3. Built for Seedance reference audio: 48 kHz stereo and at least 1.8s long, so set Minimum duration to 1.8 for a short clip. Lead silence delays the audio by that much; use it only for a generated take, to give the shot a beat before the first word. Leave it at 0, the default, for zero offset, which is what a lip sync reference needs: in a Seedance edit the audio has to line up with the clip it came from. Tail silence and the minimum pad only the end, so they never move the start. The minimum counts the lead and the tail. Free: deterministic processing, content-addressed output.
Try Audio Pad in the playground
Audio Pad
Node type: utility:audio_pad
Category: Editing
Description
Audio Pad takes one clip on Audio and returns it as Padded Audio, with optional silence before and after, padding up to a minimum length, and resampling: wire the clip in and run anything downstream of it, since the node has no Run button and refreshes when a downstream node runs. Out of the box it adds no silence and only resamples to 48 kHz stereo mp3. It is built for Seedance reference audio, which should be 48 kHz stereo and at least 1.8s long, so set Minimum duration to 1.8 for a short clip. Leave Lead silence at 0 for a lip sync reference, since any lead makes every mouth shape land late by that much; Tail silence and the minimum pad only the end. A clip already longer is never cut. Audio Pad is free, and an unchanged re-run reuses the last file.
Canvas ports
These appear as port handles on the left side of the node.
| ID | Label | Details |
|---|---|---|
audio |
Audio | AUDIO (required) |
Sidebar config
These render as form fields in the right-side config panel when the node is selected.
| ID | Label | Details |
|---|---|---|
lead |
Lead silence (s) | NUMBER |
tail |
Tail silence (s) | NUMBER |
minDuration |
Minimum duration (s; blank = none) | NUMBER |
sampleRate |
Sample rate | TEXT · options: 48000, 44100, 24000, keep |
channels |
Channels | TEXT · options: stereo, mono, keep |
format |
Format | TEXT · options: mp3, wav |
Outputs
| ID | Label | Type |
|---|---|---|
audio |
Padded Audio | AUDIO |
What it is for
Reference-audio prep in one node. Seedance's @Audio reference should be 48 kHz stereo and at least 1.8s long. Set Minimum duration to 1.8; the 48 kHz stereo default does the rest.
Leave Lead at 0, the default, when the audio is a lip sync reference. In a Seedance edit, the edited clip's own audio on Reference Audio drives the timing of the mouth, so it has to start exactly where the clip starts: any lead makes every mouth shape land late by that much. Tail and Minimum duration only pad the end, so they never move the start. A lead is for a generated take, where it gives the shot a beat before the first word.
Settings
- Lead silence (seconds, 0 to 30): silence before the audio.
- Tail silence (seconds, 0 to 30): silence after it.
- Minimum duration (seconds, blank = none): pads the end until the whole file is at least this long. It counts the lead and the tail. A clip already longer is left at its length, never cut.
- Sample rate: 48000 (default), 44100, 24000, or keep.
- Channels: stereo (default), mono, or keep. Mono sources are duplicated to both channels.
- Format: mp3 (default) or wav.
Out-of-range values are refused with the number, never clamped.
Free, and why
Free set, no Run button, refreshes when anything downstream runs. The output is content-addressed, so an unchanged re-run reuses the last file.
Auto-generated from the Wireflow node registry.