On virtual production shoots we often end up with several camera/reference angles plus separate recorder audio (dual-system sound). Getting everything lined up and into an editor is where I keep losing time, so I’m curious how others handle it.
Our current approach:
- Sync cameras and the external audio by waveform/timecode so every take lines up
- Export the synced result as XML or OTIO so the NLE (Premiere, Resolve, etc.) gets clean synced clips with audio already attached
- Then edit (or let an AI-assisted pass do a first assembly) from there
Questions for the group:
- Do you rely on timecode jam, slate/clap, or waveform matching for reference-camera takes?
- Have you hit drift or offset problems between LED-volume/ICVFX takes and the audio recorder?
- Do you prefer XML or OTIO for the hand-off, and which has been more reliable for you?
Disclosure: I’m the maker of WaveXML (https://wavexml.com/), a tool for this workflow: sync cameras and audio, then let your AI edit. Happy to hear what’s missing from your pipelines. I’m mainly comparing workflows.