I’ve been using NVIDIA MAXINE’s Gaze Redirect feature for various pre recorded videos through the run.bot method. Unfortunately, it doesn’t output the audio alongside the video, requiring me to manually reinsert the audio in my editing software. Ideally, it would be preferable if it left the audio untouched. Additionally, I’ve noticed a slight alteration in the video’s color, which isn’t a major issue since I typically adjust colors during post-production. However, the primary challenge arises with longer videos where aligning the audio’s speed with the video’s pace doesn’t always succeed. It’s peculiar that the video and audio aren’t synchronized, leading me to suspect that the video might be rendered at a different frame rate or undergoes some modification that prevents them from aligning correctly in an external editing software timeline. My temporary solution involves adjusting the video’s speed to match the audio’s duration, which suffices for shorter clips but tends to lose sync as the video lengthens. I’m searching for a method to maintain consistent video timing, as I’m uncertain about the cause of the synchronization issues with the rendered video.
Are there video output settings I can adjust or examine? Perhaps tweaking these settings could rectify the synchronization discrepancy between the original video’s length and speed versus the output.
edit: I can also confirm that the frame rate of the Nvidia gaze video is at 23 frames where the input video is at 24 frames. I’m guessing that’s the issue. Where can I adjust framerate output? Also, can I keep the audio synced in the output video from Nvidia?