Closed SeanBannister closed 7 months ago
To the extent of my knowledge and testing, having a temporary file without audio is required. There can be only one input pipe to ffmpeg of either video auto data or image data. I put a good bit of time experimenting with sending both sequentially, but could never find a workaround for the audio being interpreted as pixel data.
As for potential workarounds, it should be feasible to add a new node that would take the Filenames
output from the node and provide a number of cleanup options. I'll look into this tomorrow.
I've noticed when using the Video Combine node with an audio input it saves two copies of the video file, one with the audio and one without. Could I suggest that this should only save the video file with the audio.
If a user wants a none audio version of their video they could just use two video combine nodes one with audio and one without.
Alternatively, any suggestions on how I could automatically delete the none audio version?