Skip to main content
Mux Audio

Mux Audio · screenshot coming soon

​
What this node does

Mux Audio combines a video with a separate audio source into one clip. It takes a COMFYTV_VIDEO and a COMFYTV_AUDIO and, on ▶ Run, muxes them (PyAV / ffmpeg) into a new MP4 saved as a project snapshot. It needs both an upstream video and an upstream audio, or the run errors out. The output is COMFYTV_VIDEO.

​
When to use it

  • Drop a music bed or voice-over onto a silent (or to-be-replaced) clip.
  • Swap a clip’s original audio for a cleaned-up or dubbed track.
  • Mix generated speech / sound design over existing footage audio.

​
Parameters

​
mode

How the incoming audio meets the clip’s existing audio:
  • replace (default) — discard the clip’s original audio and use the incoming track only.
  • mix — blend the incoming track together with the clip’s original audio.

​
offset_s

Shift the incoming audio in time, in seconds. Range −600 to 600, step 0.05, default 0.0. Positive delays the audio (starts later); negative pushes it earlier. Use this to sync a voice-over or music to the picture.

​
Inputs and outputs

​
Tips

  • Both inputs are required — a missing video or audio raises an error.
  • Use replace when the source audio is unwanted; use mix to keep both (e.g. ambience plus a music bed).
  • Nudge offset_s in small steps to line up speech with lip movement.

​
Bridge note

COMFYTV_VIDEO and COMFYTV_AUDIO are project snapshot references, not native ComfyUI tensors. To move between ComfyTV and native ComfyUI nodes, insert a Bridge (ComfyTV/Bridge). See https://github.com/jtydhr88/ComfyTV/blob/main/docs/bridges.md.
  • Video Volume — set the clip’s own audio level before mixing.
  • Video Concat — assemble the sequence, then mux a single soundtrack across it.