You could repeat the pattern in the sample video file a few times, to let it calculate an average if the sync delay is varying. The virtualcam/audio device could notice this magic video frame (some never real pattern in the first line) and then look in the sound buffer for the audio spike, and calculate the sync offset. That video file would contain a distinct video image and a distinct audio waveform, and play them together. To do automatic sync, you would provide a short video file to play as a source in OBS. I can install voicemeeter and I think it can do latency, or I can add latency in the OBS advanced audio settings (which one doesn't want to do because that's global and must be turned off if you want to do a recording etc.) But one nice advantage of a simple device that was part of virtualcam is that it could include an automatic synchronizer. In followup to this, I calculated my audio as coming about 50ms before the video.
0 kommentar(er)
