Wow, another day on SL11 and I’m just blown away at what can be accomplished. All that blind NR I’ve been doing for years just mostly going out the window (thank goodness).
So I just wanted to lay out some more praise, because this program is changing my life for the better.
So far, I’m mostly working on interviews on location with a Lav and shotgun, so layers are seldom over 3 or 4 in a finished job. It just takes a while to get a firm grasp on what you are looking at to edit…similar to amplitude waveforms. Work with that long enough and you can begin to recognize letters in speech.
I was telling a visitor about SL last week…his comment was: “sounds like a glorified EQ”…ok, I guess so…nevertheless, I don’t see ProQ3 capable of what I’ve been doing with SL!
I’m using plenty of Unmix Noisy Speech and Transfer tool for my purposes.
So, thanks again to @Robin_Lobel for this incredible tool! And all the other forum members responding to my posts; your advice has been extremely helpful to me.
5 Likes
… and birdsong … and jackhammers … and … and … 
2 Likes
I did mean in waveform display form 
yeah, spectral view taking it much farther
I got birdsong to work on tomorrow
holy smokes! another day with SL11…delivering me amazing results
Looking forward to tomorrow
I’m certainly considering mixing as well while in the Spectral editor
thanks to @Phil_Pendlebury latest tutorial showing me clever use of EQ Matching, too!
3 Likes
This is great to hear. Thank you for posting back.
2 Likes
Oh man, I feel like I’m really getting the hang of SL manual editing…I suppose working for hours with it every day, You gain a workflow which leads to some excellent results. I’m really enjoying this…
2 Likes
ok, well, I find Harmonic selection works great on singing; not so good on dialogue (not at all on my current project (french dialog/IVs)…so I’m carving the harmonics out manually with mostly transfer tool…very time consuming.
Again, a processes discussion would be great 
My process is:
Unmix Noisy Speech equals two layers;
then,
create another layer for wanted and unwanted noise
I’m finding it much easier to transfer voice out of the noise or other layer or wherever there are remnants…it’s just much easier to see if the voice layer is visible and just turn down the volume to adjust opacity. I can foresee really wanting to have 2 active layers for this purpose in future.
I’m not using any destructive processes (erase, NR)…so, really I should be using another layer for voice remnants so I can back out if I needed…because transferring the remnants back to the Speech/Vocal layer is destructive.
1 Like
OK, I’ll keeping documenting my process in the event someone who might be searching online might find helpful…I would have 
So, rather than the time-consuming activity of carving the speech remnants out of unmixed noise of Unmix Noisy Speech with just the Transfer tool, the Harmonic Selection tool can save a lot of time.
First, ensure layers are best placed; add a layer below the Unmixed Noise layer if needed to cut/paste or transfer the remnants;
then,
Using the Speech layer to set the selection of speech with Harmonic Selection tool>
highlight the layer from which you want to remove/reduce the remnants>
Shift+X to cut special to layer below
Then repeat or switch to carving out the remnants with the transfer tool
Previously I was trying to use the Harmonic Selection tool on the Unmixed Noise layer, which pretty much won’t select the correct harmonics.
Simple, but I’m learning as I go.
1 Like
It’s really great fun to discover the possibilities of SL!
Have you tried Select › Invert Selection Frequencies while playing with Harmonics selection? It makes it easy to get rid of mud ’between the lines’.
3 Likes
The more I work with SL, the more I recognize useful tools in the displays
For example, the waveform display…I presumed I wouldn’t need it…BUT, I’ve been staring at amplitude waveform displays for a long, long time…and it essentially shows the amplitude cross-section…hence having the waveform in the center of the spectrograph…really, the better you are knowing what you see in the displays, the easier to arrive at good results
yep, finally nailing down a workflow for my Lav and Shotgun recordings of mainly dialogue…amazing power with SL11…Transfer tool is an absolute must
Lavs have turned out to be a fair bit more work than expected, I’m essentially doing:
Unmix Noisy Speech
then;
Unmix Levels on the Speech layer (-65dB)>Hi @ unity/Lo ~-9dB
then manually transferring from Lo to Hi layer to bring out buried (mostly) high freqs
also going thru the Unmixed Noise for wanted inhales, sighs, wanted rustle etc…
and, of course using an Unwanted Noise layer rather than destructively remove unwanted racket…(helps show my work to producer… director, etc)
I tried Unmix Layers as a first process, but that wasn’t working for me
With Shotgun, I want more ambience,
so Unmix Noisy Speech
then go through the Unmixed Noise and separate into Wanted and Unwanted layers
I’m getting far fewer artefacts than my previous years of using NR VSTs along with better sound, hooray!!! I am having to remix transitions of my comps before putting into SL because I was previously using old school room tone cover processes. With SL I can use what the director and editor cut. Crew chatter is pretty easy to cut out, for example.
2 Likes
do you do a simple cleanup/get-it-together-edit first, then bounce it and bring it into SL standalone?
1 Like
OK, more than that…I’ve been working on this film for about 5 years!
So, I have the film’s dialogue transitions all mixed already
With my comps, I have Acon DeClick and FabFilter EQ and DS already burned in when I export…I could bypass it, but I really just need NR at this point. I will have to EQ again in DAW…But now I will have very few FX in DAW when I remix. I’m having to remix all the cover before bouncing out of DAW.
Yes, I’m 100% standalone with SL
Yeah, what happened is the first 30min was different mixing style, which we changed on the last hour. We essentially decided to do hard cuts with picture rather than smoothing transitions. Well, SL helps even more with that 
The biggest deal is getting rid of blanket Noise Rdx which has so many artefacts depending upon scene…Spectral editing can drastically reduce those artefacts IME
Wow, I’m getting better with this fantastic software every day
now it’s time to dig some voices out of un-mic’d contributors who were passing by …I’m looking forward to this…I’ve certainly got a method to deal with the TX/RX blasts on the radio mic…takes a little work, but the results are so impressive
2 Likes
workflow finally sinking in
I hadn’t tried flipping phase to deal with crew chatter and that can sometimes work wonders!
Also, getting down what layers you want to separate which audio into…and how to color those layers is critical…I’ve got my scheme that works for me
we are amazed at what can be accomplished with SL every day