Hi everyone. I’ve been lurking for a bit, but this is my first post. I’m a data engineer, so I’m used to wrangling data pipelines, not video files. I just got tasked with editing a multi-camera Zoom recording for a company webinar, and I’m honestly a bit terrified of messing it up.
I downloaded Descript because I heard it’s good for podcasters and it has this transcript-based editing. But I’m starting with a Zoom recording that has:
- One main video file (Speaker View) with everyone in grid
- Separate audio-only files for each participant from the original Zoom cloud recording
I’ve imported everything, but I’m immediately overwhelmed. My main questions are:
1. Should I be combining these into a multi-track timeline first, or is there a smarter way to sync them using the transcript?
2. I’m worried about overwriting or losing the original sync. Is there a “non-destructive” or safe pattern here, like how we version control SQL scripts?
I saw the multicam editing tutorial, but it assumes you start with separate video files. Mine is mostly one video with multiple audio isolations. My goal is to clean up pauses, ums, and maybe switch to a specific speaker’s audio if the main track has cross-talk.
Any step-by-step guidance or a “data pipeline” mindset for this in Descript would be a huge help. I don’t want to break production, even if it’s just a video this time.
Sync is your primary concern with a setup like that. Your one video file with multiple audio tracks is the baseline. Drop that into Descript's timeline first.
Those separate participant audio files are your insurance. Don't combine them on the timeline yet. Use them to replace the audio from the main video file if someone's mic was bad. Descript's transcript will be generated from the main video's audio track, so any replacement you do later will need a manual check for sync.
Version control: just duplicate the Descript project file before any major edit. Save each iteration with a date or version suffix. It's crude but it works.