I’m doing some dialog editing and I find the workflow I’m using confusing;
I’m editing smallish events that relate to longer source files (take 1, take 2, sinchro clap, etc) when I set SL as ARA extension for those clips, they are each added to their own layer. Thus I end up with 50 to 100 layers a lot having the same name (the source file name).
When I double click the event for editing in SL the zoom is all over the place ex: the event I just clicked is scruncht up to the left side and not practically editable ( I also have events overlaping each other) so in order to perform the required SL editing I have to:
solo the desired Layer (many have the same name) +
select the Zoom icon +
click Zoom Active Layer to have a workable view +
pull down the wave/spectrum divider to display the waveform (hopefully only once)
So my question is : Is there an easier way to do dialog editing in SL.
Regards,
Marius
P.S. You guys are the best and have always come up with a solution or answer
I work in standalone…it can be a bit tedious; but I certainly have control of naming everything and don’t have concerns about funky relationships between programs…I find the reliability helps me chip away through the spectral editing
I’ve grappled with this too. Unfortunately I haven’t come up with a solution that works the way I expect it to. Having dozens of layers in SL becomes untenable ( for me anyway ). The 'Group By…" display function in SL at least brings all the media into one view. But it too gets confusing.
My solution has been to work on one event at a time ( save a module chain in SL if you want to able a process to many events which require the same processing ) and then use “Make Extension Permanent” . This stops the layers building up . Yes, I hear you take a breath at the idea of committing to a SL process. You could duplicate all your source files and use SL on the copies?
Additionally I always try the “Voice Isolation” in Nuendo before SL . It’s not bad at all!
I hope SB work to integrate SL in a better way than ARA. I’d like to be able to select a bunch of events in Nuendo and have them open in their own SL Project. When I save the the SL project it appears as a single event in Nuendo with the rendered SL output. Later, double clicking the SL Event in Nuendo opens the SL project again, with its undo buffer in place.
At least I’m not the only one struggling with ARA workflow and integration. Big INHALE, yes “Make Permanent” scares me too, it’s goes against the whole principle of non-destructive editing. But I guess it’s the only way to go. I’ll be using “Duplicate Track” a lot it seems.
Steinberg has done amazing work with SL and it’s algorithms…it’s always the integration and workflow that takes more time
Hi everyone, I’ve really enjoyed following this discussion — lots of helpful insights here!
Like many of you, I also find that ARA-based dialogue editing often results in a rather complex workflow.
Especially when working with a large number of events, managing layers and keeping things efficient can be a real challenge. I hope this area becomes more streamlined in the future.
Here’s a look at my current workflow:
I start with light noise cleanup using iZotope RX via Direct Offline Processing, with a 5-second handle added.
Then I open the event in SpectraLayers via ARA Extension.
(Note: Since ARA doesn’t support handle regions, I use a macro to extend the event by 5 seconds beforehand. Also, if the event overlaps with a neighboring clip, it won’t transfer into SpectraLayers correctly via ARA.)
In SpectraLayers, I perform multiple separation passes to isolate dialogue and noise with high precision.
To avoid bloated sessions from too many active ARA links, I use “Make Extension Permanent” to commit changes.
If I want to reuse the separated audio, I drag and drop the layer back into the Nuendo Edit Window to create a new event.
Due to a known bug that sometimes introduces noise above 15kHz after ARA processing, I run another RX cleanup via DOP, and finalize it with “Make Direct Offline Processing Permanent.”
(Since this is destructive, I use it carefully.)
Because this workflow doesn’t preserve the original event, I always keep a duplicate of the original clip on a separate track, aligned to the same timeline, for safety.
SpectraLayers’ separation algorithm is truly excellent.
If the ARA integration and UI for working across multiple events improves, this would become the ideal tool for film dialogue work.
Robin (the developer) mentioned that multi-event improvements are coming in SL12, so I’m really looking forward to what’s next!
Great to hear that SL 12 will include some workflow considerations
Being old I recall that everything in post production used to be destructive ( cut the tape or film, record over the old stuff with video etc ) so we’ve been spoiled with digital editing and the ability to ‘go back’ and start again! So “Make Extension Permanent” isn’t really such a hardship. The original file is always in the Pool ( somewhere… ) but to avoid frustration, my process always includes an industrial level project sorting/organizing /naming stage right off the blocks, That way I can duplicate everything into a single muted “group folder” so that anything can be quickly copied and pasted back to the original state. Not that I didn’t properly organize before… but you know, sometimes I’d get going before everything was completely tidy!
This is a great thread! Thanks ltf3 and nabw. My background is Computer Programming and like ltf3 I like everything properly named and organized. But since I’m dealing with Premiere Pro AAFs and in this particular project, the editor Merged the audio files with the sequence and at AAF export the file names are randomly generated (ex: 9e3d65c3-2e93-9094-5856-ba1a00000083) such that it is cumbersome to reference the sources. I need to find a way to batch bounce all of the files at the start of the Dialog Editing process so as to avoid all those “New Version” popups every time I do a destructive edit. Any ideas ?
BTW, this is my first Dialog Editing project. It is my daughter’s first feature film so we’re both new at this (explains the merging instead of using “multi-cam” ) and I’m still learning Cubase features so any tips for “batch bouncing” the events in Cubase are welcome.
