Cutting subclips with a parallel soundtrack

Hello all, I am trying to edit film that has two soundtracks, one native language (associated with the footage) and one separate track that is dubbed. In the timeline, the dubbed track lines up perfectly with the native track (all sfx and music is exactly in line), so I have no issues with that. But if I put the film (and its associated native-language audio) in the source viewer to cut subclips, can I set the dubbed track parallel with the native-language so I can get identical subclips from it? If this is a silly question, well, it's not the first time I've asked one... This newbie will thank anyone for answering!

regards,
John Tenney

I think the only way is if all video and audio tracks are part of one file.

Thanks Jim. Can you group all the tracks together into one "unit" that can be edited together?

Into subclips? Not that I'm aware.

Or maybe, rather than into subclips directly, into a "meta-track" that can be cut into subclips with all the component parts reflecting the same in and out points.

I'm not an audio guy, and I've only just begun using multichannel audio much, but I think you'll want to try combining both soundtrack versions into one file, using a container format which supports multichannel audio. Then you should be able to edit them in tandem, and export them selectively. .Mov/prores will work if you're on a mac. Not sure of a Windows format, but if I had to guess, I'd try DNx.

Does this sound right to the editors here with localization experience?

Or maybe, rather than into subclips directly, into a "meta-track" that can be cut into subclips with all the component parts reflecting the same in and out points.

You could use multicams. Actually, if you're a new editor, multicams might be a more intuitive place to start than multichannel audio.

Thanks Jason! Actually, I am an audio guy, coming to (better) video from a background in Apple Logic Pro. At this time I am using H.264 clips from multiple online sources. Hardly professional... but that's the extent of my visual projects for now. So I don't think multicam will be applicable to what I'm trying to do. And "container format which supports multichannel audio" goes over my head I'm afraid. I've been trying to research all this stuff, but because of the lack of indexing in the reference material I'm reduced to searching text strings in the pdfs, which isn't ideal...

So I don't think multicam will be applicable to what I'm trying to do.

Well, I am probably making some implicit assumptions about what you're trying to do, so I might be on the wrong track, but I also haven't given you a very detailed explanation about how you might use multicams to get what *I think* you're after, either. You can use them for more than just switching between different camera angles. You could also use them to switch between different audio "angles," which could be just what you need to do, depending on your situation.

Here's what I'm a little fuzzy on: When you say that "I am trying to edit a film that has two soundtracks, one native language (associated with the footage) and one separate track that is dubbed," it sounds to me like you are editing backwards. Like you're deconstructing something. To me, "films" don't have "soundtracks" until they've been in theaters and made deals with major record labels to put CDs in jewel cases to sell next to posters and merch. So, guided by this logic, I might assume that you're taking a completed work, such as a DVD with different language localizations, ripping it and all of its constituent languages, and chopping it up in Resolve to re-mix into a new-and-protected-by-fair-use-derivative-work. Which is totally fine, nobody's here to judge, bro ;-}

On the other hand, taking "I am trying to edit a film that has two soundtracks" at face value could mean that you're collating a heretofore incomplete motion picture from thousands of pieces of original media. That you have a massively complex timeline, stacked with layers upon layers of edits, and that you've obtained the second language localization through hours of ADR.

Now, I *know* that each of these guesses is a million miles off base, but here's my point: In the first scenario, you're likely dealing with two audio tracks. In the second, you could be talking about hundreds. Hundreds of tracks and complicated timelines won't really "nest" well into a multicam, but two stereo tracks will. And if all you need is a "meta-track" from which you can pull short clips, each with ins and outs synced across two versions of temporally aligned audio, then multicams should work perfectly.

I've been trying to research all this stuff, but because of the lack of indexing in the reference material I'm reduced to searching text strings in the pdfs, which isn't ideal...

Yeesh. I hear ya on that one, brother. The Fusion stuff is a nightmare.

into a "meta-track" that can be cut into subclips

Sure. Export a new file.

Thanks for your response Jason! The forum site lost a longish response I wrote, trying to be more specific, so I will rewrite it and save it before sending it off. Damn...

I am not trying to do your second possibility. It is much simpler. I took a film from YouTube with a soundtrack containing English dialogue. Then I got audio-only of the same film with a soundtrack dubbed with German dialogue. In the timeline I lined the German audio-only up parallel to the audio of the English-language film, and was pleased that the sfx and music cues lined up exactly, so that theoretically I could switch back and forth between languages, which is what I want to do. I want to create subclips where potentially I can work parallel in the video and both audio tracks, since they are so nicely lined up. I suspect I am trying to run when I can't walk yet. My video skills are far behind my audio, which is one reason I came to DR in the first place, and I've definitely made some progress.

The DR community is so helpful! It gives a newbie like me a lot of confidence going forward. Thanks for taking the time, Jason!

best,
John

Cool. Ok. Now we're getting somewhere.

So, probably the easiest way to proceed, since it sounds like you already have everything lined up on the timeline, is this:

1 - Select everything on the timeline that you've manually sync'd by drawing lasso around it. Include both audio and video.
2 - Right click on any one of the selected clips in the timeline, and choose "New Compound Clip..." from the contextual menu.

That's about it. You'll notice that everything you had previously selected collapses down into a single video track and a single audio track, which has a new name (probably "Compound Clip 1" unless you changed it). This is a nondestructive container which houses your original elements just as they were when you invoked the command. To prove it to yourself, you can break the original elements back out of their container by right-clicking on the Compound Clip in the timeline, and selecting "Decompose in place." The container then pops open, exposing your original elements.

"But hang on," you say. "I wanted to select INs and OUTs in the source monitor and populate my timeline with little bilingual baby clips."

Fear not. You can still do that. In fact, even if you now delete all of the clips on your timeline, you can still get back everything you'd originally selected when you created the compound clip. Because: If you'll notice, there's now an object in your media pool called "Compound Clip 1" or whatever you named it.

When you create a compound clip, you'll notice two things happen: The more obvious effect happens on the timeline, where all of your clips get collapsed into a container, which you can blade, trim, and manipulate like any other clip. But the less obvious effect is that the state of every selected element gets saved into a new entity, which appears in the media pool, and is called a Compound Clip (henceforth abbreviated). You can think of this as the CC, proper, or the über CC, or Master CC, or whatever you want. But, in fact, the A/V container strip CC you see on the timeline, as well as the one in the media pool are all instantiations of each other. If you right-click on any of them and select, "Open in timeline," you'll be transported "inside" the über version, to the original arrangement of your normal clips the way they were when you invoked "New Compound Clip..." When you're inside this special place, you can edit the original arrangement, and then pop back out again by double-clicking the "breadcrumb" text at the bottom-left-hand side of the timeline pane (Hint: if you left all names at default, it'll say "Timeline 1 > Compound Clip 1". Double click on "Timeline 1" to jump out.) When you jump out, every instance will reflect the changes you made inside.

"But my bilingual baby clips?"

Yeesh. Calm down, I'm getting there.

So, understanding the way the instantiation works is important because you want to be able to control de-instantiation. Let's use your YouTube project as an example. Let's say you've got your video clip on V1, the English audio on A1, and the German on A2, and everything's sync'd. You lasso everything and create CC1. You blade CC1 into thousands of bits, shuffle everything around, using the source viewer (if you want), timeline trim tools, and whatnot. Then, you decide that the German audio is a little too quiet. No problem, you jump into CC1 by saying "Open in timeline," raise the clip volume level on A2, pop out via the breadcrumbs, and your changes propagate to all thousand clips. Easy peasy.

But then you decide to tackle a different issue: this whole time, you've been hearing both the German and English mixed together, and that's not what you want. Let's pretend you want to hear English at every 1of3 clips, German at every 2of3, and silence at every 3of3. So, you try popping into clip 1, turn on the english, turn off the German, pop back out, and everything reflects the changes you've made. Everything is in English at this point. If you try the same trick at clip two, when you step back out, everything will be in German, including clip 1, which you already painstakingly altered. So what do you do? You de-instantiate by turning the CC's back into normal clips. And how do you do that? Select all the CC's you want to de-instantiate, right-click, and select "decompose in place."

So the steps here would be:
-Pop in to the CC, adjust until all you hear is English, pop out to master timeline
-Select clips 1,4,7,10,13,[...], "decompose in place"
-Pop back into the CC, adjust until all you hear is German, pop back out
-Select clips 2,5,8,11,14,[...], "decompose in place"
-Pop in, make silent, pop out
-done

This type of workflow works when you have a really long timeline, with lots of edits, and only a few different states to switch between. English/German muting and unmuting is a perfect example of a simple, alternating, categorical change, to which the workflow is well-suited. But as your adjustments require more granularity, the more you'll need to consider a different approach. Often, you paint the broad brushstrokes with a method like this, then go over top of everything manually with the spit and polish.

Hope this helps.

Oh, and I know I hyped up Multicam earlier, and this post is all about Compound Clips. Multicam Clips are special compound clips with extra superpowers. Those superpowers might help a little in your case, but the extra learning overhead isn't worth it until you understand Compound Clips, because they will also work for what you need.

What you may want is to make a multicam file, which is a basic way to ingest you extra audio file into a container that also includes the main video/audio clip. Even if multicam has another purpose, this may help...
So, select 2 clips (any) which have audio > right-click > make a multicam file > right-click this multicam file > "open in timeline" > wipe out the 2 files... That was just an easy way of creating the multicam container (named ISO in Resolve)

Now copy your aligned tracks - with desired audio ones - you had made on another timeline and paste everything here, on this very multicam open timeline, and double-click on the bottom left name to close it.
Nox you have a single file with both of your audio tracks inside (hoping this is what you wanted )

Hello,

Another solution. Much less, much less elegant than multicam but it works.

You link all tracks between them (select all tracks and right click => link). Your first clip now has 2 audios (native + dubbed)

You open a multi TL and by I / O you copy / paste in the bottom track. The result is a source TL and a destination TL at the bottom.

Cutting subclips with a parallel soundtrack.jpg

Thanks so much, Jason, Sam and Jean Claude. I have been away from this forum, and Resolve, and am now going to reapply myself to your solutions. It is interesting, and very cool, that Resolve is flexible enough that a problem like mine can invite three such different solutions. Very much obliged, gentlemen!

Now that some of "life's dust" has settled, I can report on my solution to my problem. I think I solved it, and it seems somewhat simpler than the solutions from Jason, Sam and Jean Claude.

All I had to do was establish a workflow based on timecode. Since the two soundtracks were absolutely in sync--just with the different languages in the vocal track--I was able to 1) pull the video/audio 1 track into the source viewer, 2) do the usual in/out, paying close attention to the timecode, 3) create a subclip and change the name in the media pool if necessary, 4) bring the audio 2 track into the source viewer, 5) in/out using the timecode as reference points, 6) create a subclip of that and rename if necessary. Then, as long as I keep the naming consistent, I can easily line up the three elements--video, audio 1 and audio 2--and go from there importing into a timeline as needed.

This may not be very elegant, but it works for me, and I think I can live with it.

Thanks again to Jason, Sam and Jean Claude for the time they took to get me sorted out!

from John, the grateful newbie

Somewhat simpler