VoiceOver / Scene mental model
J
Joel Kang
Voiceovers shouldn't be restricted to a single scene. I want to span my voiceover across multiple scenes / pieces of media. When i resize a scene to be shorter than the last voiceover item, it removes the voiceover entirely instead of adding it to some "unplaced" bucket
Log In
Derek Feehrer
Yeah this is a tricky one. So originally voiceovers were global, and lived alongside music outside of any scene. There were a few issues:
- It was hard to line up voiceovers with text, without being able to see them next to each other in the scene timeline. In many videos voiceovers and text are closely related and need to be synced.
- Any time scenes were reordered, extended or shortened, the voiceovers would move around relative to the scenes, since they weren't attached to the scenes in any way. This proved quite frustrating.
Adding them to the scenes made it a lot easier to use voiceovers in most cases for most people. But I agree completely: 1. there are cases where you want voiceovers to span multiple scenes, and 2. the deletion on resize is annoying. An unplaced bucket is a good idea and we can totally look at that.
For spanning multiple scenes, one option is a global toggle that would let you choose if you want to use voiceovers inside scenes or globally. Would be fairly easy to implement but not sure how intuitive it would be.
Any other ideas? What do you think would make sense Joel Kang?
J
Joel Kang
Derek Feehrer I actually think the main crux of the problem is the concept of the scene -- right now a scene is a 1:1 association with some media, but that might not be true (hence also the other request to have multiple clips per scene).
In my opinion, a scene is more of a logical grouping of mostly-related things. In the common case, indeed the visuals, VO and text
only
relate to each other and not to any other scene. So I can see how a mental model where you're working at the scene-level seems intuitive. But if were to allow multiple clips per scene, or allow VOs to span multiple scenes, then really what
is
a scene (since I can have my entire video be 1 scene). The fact that I can also split a scene into multiple scenes also proves my point that scenes actually just logical split points -- except now i have a shorter window to force my VO into. I think that's why traditional video editing software don't typically force you to work at a scene-level, and instead avail the entire timeline for you to scrub along. I'm sure y'all must have thought long and hard about the scene-based mental model, so there must be things that I haven't considered.