Being an avid RX user for years I demo-ed SL10 a few weeks ago to see if (like you say) it’s an alternative for the Izotope app. I found the quality of some of the functionalities lacking in SL10 compared to RX10. RX somehow delivers a more precise result and offers more parameters to tweak. So I’d say it’s not an alternative but definitely adds to the restoration and mastering toolbox. 😊
I'll agree - especially when you spent years with RX then the switch isn't easy. SL11 though improved the user interface. But what really annoys me with RX is that they haven't improved their interface in years. I can only speak for my workflow but its a lot about creating regions and selections. And that functionality is implemented absolutely weak. You can't lock a selection on the axis when moving around (like any layout software does since more then 30 years). You cannot select several regions from the list and activate the selection. Some tools still only work up to stereo channels and can't be used in multichannels like 5.1. There is no GPU acceleration. Transciption feature is so bad it's laughable and still only works in English. Wave plugins do not work in RX (but you can load them via SigMod from Nugen). The last version updates had almost no new features you would use for the core functionality - restoration and cleaning up.
Hi - I am trying to separate kick, snare, hats and cymbals from a drum track that was done live many years ago... it's my session, I am the sax player. I have the trial of this app, but I cannot figure out how to get past the "unmix" part. How do you download the separated kick, etc? Or can this software not do that? Thank you!
I havent and it seems weird for sure, as if the files get rendered to 44,1 when you work in 48khz. I would go to steinberg forums to see if more have that issue.
@@audiotoolshed Thanks for the answer man! - I tried in the forums no luck yet - After I Unmix a song the Bpm speeds up & also the audio doesn't match the visual wave of the event.... - Is it something with latency? Do you have any idea what I should try? (I work on Cubase13)
I demoed Spectral Layers 10 a few Months ago in My Studio One 6 and I thought it worked great! I still do not like the high costs of some of this Restoration software plugins, but I like what they do. Great video White Noise
I feel like I am being dumb, but I can't help but wonder why Steinberg waits about a month post-release before releasing a demo/trial version of SL (as opposed to a trial being available alongside the full version's release).
I've got a Nvidia RTX A6000 with 48GB GPU memory (it's equivalent to a RTX 3090 but double the memory). With GPU in use it takes 5:18 min to unmix a 2:27 min song at extreme settings and 2:15 min at high settings. But monitoring the process with ProcessExplorer it looks like SpectraLayers 11 does top out at using 3.5GB of GPU memory and GPU usage jumps up and down averaging at 20%. To me it looks like there is room for improvement and maybe they didn't expect a GPU with that much RAM or did restrict the memory usage and crippled performance by that.
I can confirm that GPU behaviour, i did a little test yesterday with a regular 3080. GPU spikes up and down with that card to 29% percent, while the CPU runs at 42-ish percent and that is with GPU rendering enabled in spectralayers. Almos looks like it reads / computes and then writes a bit everytime.
@@audiotoolshed It also takes >20 sec of freezing after reaching 99% - until it is finally showing the unmixed layers. Have also tried several VST3 plugins and some of them are off-setting the audio to the left after rendering - for example RX10 and Ozone10 plugins. Gullfoss plugin did stay in time. The whole interface of SL11 feels easier to use now.
Unless you activated it after 15th of May, I wouldn't count on it..... I bought it in March or April. Activated it back then. Now I've got a new computer, my last activation was May 20th, but I think I'll still need to pay for the update (it's not that much €67)
@@Henry0870 As you mention, May 15th is the date according the forum too, so I'm out and will have to buy it. But you may have a voucher for your second activation, let us know 🍀
Thanks for this! This is the best vid I have seen on deep usage of SL so far. I learned new things here immediately. My main usages are in post for film and unmixing of original music from the 80s and 90s. I'm looking for improvements in workflow and you certainly provided some of that here! I've known about and trialled Spectral Layers in the Sony days and just didn't previously have a computer powerful enough to run it. I've been working on a feature length documentary and had need of forensic audio separation and NR and bought SL10, so I'm kind of new to SL and looking for ways to manipulate audio SL beyond how I approach it. I haven't seen any vids explaining editing through separated "unmix noisy speech" and then editing the noise to wanted and unwanted noise...maybe that's too boring for most folks? I've been using Acon Restoration Suite and DeVerberate and Sonnox Restore for years, but SL is giving so much more control and as much as I don't care to bounce, SL has changed my workflow in that regard. I'm also seriously considering changing DAW (I've been using SAW Studio since the early 2000s) and your showing that SL can work within Cubase with video is a question I've been wondering about. Is your video running in Cubase natively? Or does that require an add-on/ 3rd party or particular workaround? Or is Nuendo required for native video? Again, great video!
Thanks! Cubase plays back 1 video file - but you sometimes have to make sure its the right format for smoothest playback. depends a bit on what system you run. mp4 runs fine for me, lower resolutions run better, but idealy you want someting like prores to get frame by frame accuracy which mnp4 doesnt have.
Biggest plus point for the upgrade is the ability to drag and drop multiple/all stems at once into Cubase. (I'm still new to SL 10, so maybe I missed it in 10??)
@@audiotoolshed Cubase. I just started using Spectralayers 10. Planning on "remixing"/remastering an album I did in the late 90's, before uploading it to Spotify and such. But it's 6 or 7 stems per songs, times 27 (songs...) I did 1 or 2 songs, but will start over and redo the unmixing with SL 11, as the quality also seems to be better in 11 vs 10.😎
i would definitely NOT recommend buying this. the unmix module can't even separate out the most basic stems. if you try to unmix a basic pop song, spectra won't be able to discern between a kick drum and a snare drum. it's just a giant unusable disaster - same as all unmix algorithms before it. nothing groundbreaking here unfortunately, and now i spent 300 bucks on this useless software because there was no trial... feels like that was probably by design. if you're looking for a way to separate stems from a song, this aint it.
![Felix "Unfair" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/Oswujxm2Ag0/mqdefault.jpg)
Being an avid RX user for years I demo-ed SL10 a few weeks ago to see if (like you say) it’s an alternative for the Izotope app. I found the quality of some of the functionalities lacking in SL10 compared to RX10. RX somehow delivers a more precise result and offers more parameters to tweak. So I’d say it’s not an alternative but definitely adds to the restoration and mastering toolbox. 😊
I'll agree - especially when you spent years with RX then the switch isn't easy. SL11 though improved the user interface.
But what really annoys me with RX is that they haven't improved their interface in years. I can only speak for my workflow but its a lot about creating regions and selections. And that functionality is implemented absolutely weak. You can't lock a selection on the axis when moving around (like any layout software does since more then 30 years). You cannot select several regions from the list and activate the selection. Some tools still only work up to stereo channels and can't be used in multichannels like 5.1. There is no GPU acceleration. Transciption feature is so bad it's laughable and still only works in English. Wave plugins do not work in RX (but you can load them via SigMod from Nugen).
The last version updates had almost no new features you would use for the core functionality - restoration and cleaning up.
Hi - I am trying to separate kick, snare, hats and cymbals from a drum track that was done live many years ago... it's my session, I am the sax player. I have the trial of this app, but I cannot figure out how to get past the "unmix" part. How do you download the separated kick, etc? Or can this software not do that? Thank you!
Can someone help? - I recently upgraded to SL 11 & now everytime I Unmix a song it speeds up the tempo... Anyone encounter this problem?
I havent and it seems weird for sure, as if the files get rendered to 44,1 when you work in 48khz. I would go to steinberg forums to see if more have that issue.
@@audiotoolshed Thanks for the answer man! - I tried in the forums no luck yet - After I Unmix a song the Bpm speeds up & also the audio doesn't match the visual wave of the event.... - Is it something with latency? Do you have any idea what I should try? (I work on Cubase13)
all that I need, is just to separate from audio, the vocals and piano, into midi, is there anything, that's accurate? maybe chord changes would help.
Can we open this in fcpx? I have spectral Rx 10 but it doesn't open in fcpx. Thanks
I demoed Spectral Layers 10 a few Months ago in My Studio One 6 and I thought it worked great! I still do not like the high costs of some of this Restoration software plugins, but I like what they do. Great video White Noise
Thanks! For me it’s invaluable for how much I use it.
High costs? nope. Spectralayers is cheap compared with iZotope RX or other audio restauration competitors...
Will this work with Ableton and fcpx? That's what I use. Thanks. Great videos. Keep em coming
regarding the long processing times: If you have a decent graphics Card you can Switch SL to use the GPU instead of the CPU. This can bei MUCH faster!
Great tip!
I feel like I am being dumb, but I can't help but wonder why Steinberg waits about a month post-release before releasing a demo/trial version of SL (as opposed to a trial being available alongside the full version's release).
There’s one available, but it’s version 10 don’t know why and I don’t understand that either.
@@audiotoolshed It seems so stupid to me, delaying people's ability to try out and potentially buy SL.
I've got a Nvidia RTX A6000 with 48GB GPU memory (it's equivalent to a RTX 3090 but double the memory).
With GPU in use it takes 5:18 min to unmix a 2:27 min song at extreme settings and 2:15 min at high settings. But monitoring the process with ProcessExplorer it looks like SpectraLayers 11 does top out at using 3.5GB of GPU memory and GPU usage jumps up and down averaging at 20%. To me it looks like there is room for improvement and maybe they didn't expect a GPU with that much RAM or did restrict the memory usage and crippled performance by that.
I can confirm that GPU behaviour, i did a little test yesterday with a regular 3080. GPU spikes up and down with that card to 29% percent, while the CPU runs at 42-ish percent and that is with GPU rendering enabled in spectralayers. Almos looks like it reads / computes and then writes a bit everytime.
@@audiotoolshed It also takes >20 sec of freezing after reaching 99% - until it is finally showing the unmixed layers.
Have also tried several VST3 plugins and some of them are off-setting the audio to the left after rendering - for example RX10 and Ozone10 plugins. Gullfoss plugin did stay in time. The whole interface of SL11 feels easier to use now.
yup. I assume thats a sort of final compsing. That offsetting sounds like a latency compensation bug.
@@audiotoolshed Steinberg released an update and addressed what we reported here - impressive. Haven't found time to play around with the update yet.
@robertniessner554 yeah Robin is usually on top of it.
Great video (as alawys 👍)! Now I know I NEED to upgrade.
I bought SpectraLayers Last October... I hope I get a Voucher from Steinberg
Unless you activated it after 15th of May, I wouldn't count on it.....
I bought it in March or April. Activated it back then. Now I've got a new computer, my last activation was May 20th, but I think I'll still need to pay for the update (it's not that much €67)
@@Henry0870 As you mention, May 15th is the date according the forum too, so I'm out and will have to buy it. But you may have a voucher for your second activation, let us know 🍀
@@MarcoPolux Nope, I had to buy the update……
Thanks for this! This is the best vid I have seen on deep usage of SL so far. I learned new things here immediately. My main usages are in post for film and unmixing of original music from the 80s and 90s. I'm looking for improvements in workflow and you certainly provided some of that here! I've known about and trialled Spectral Layers in the Sony days and just didn't previously have a computer powerful enough to run it. I've been working on a feature length documentary and had need of forensic audio separation and NR and bought SL10, so I'm kind of new to SL and looking for ways to manipulate audio SL beyond how I approach it. I haven't seen any vids explaining editing through separated "unmix noisy speech" and then editing the noise to wanted and unwanted noise...maybe that's too boring for most folks?
I've been using Acon Restoration Suite and DeVerberate and Sonnox Restore for years, but SL is giving so much more control and as much as I don't care to bounce, SL has changed my workflow in that regard.
I'm also seriously considering changing DAW (I've been using SAW Studio since the early 2000s) and your showing that SL can work within Cubase with video is a question I've been wondering about. Is your video running in Cubase natively? Or does that require an add-on/ 3rd party or particular workaround? Or is Nuendo required for native video?
Again, great video!
Thanks! Cubase plays back 1 video file - but you sometimes have to make sure its the right format for smoothest playback. depends a bit on what system you run. mp4 runs fine for me, lower resolutions run better, but idealy you want someting like prores to get frame by frame accuracy which mnp4 doesnt have.
@@audiotoolshed thanks, well, I typically just hound my video editor to get me the functional version :-)
Biggest plus point for the upgrade is the ability to drag and drop multiple/all stems at once into Cubase. (I'm still new to SL 10, so maybe I missed it in 10??)
its extremely convenient. Which DAW do you use?
@@audiotoolshed Cubase. I just started using Spectralayers 10. Planning on "remixing"/remastering an album I did in the late 90's, before uploading it to Spotify and such. But it's 6 or 7 stems per songs, times 27 (songs...) I did 1 or 2 songs, but will start over and redo the unmixing with SL 11, as the quality also seems to be better in 11 vs 10.😎
@Henry0870 drag and drop should be available in 10 with cubase if I’m correct.
I wish you lots of luck and patience with 27 songs of unmixing ;-)
Drag and drop of multiple layers at once is indeed new to 11 :)
@rldivide Thanks Robin, of all people you should know ;-)
definitely like the upgrade so far
Same, have been running it for a while now ( beta tester here) . Very proper upgrade for sure.
Is there a Demo?
yes sir: www.steinberg.net/spectralayers/trial/ but its still on version 10 apparently
Thank you!!
i would definitely NOT recommend buying this. the unmix module can't even separate out the most basic stems. if you try to unmix a basic pop song, spectra won't be able to discern between a kick drum and a snare drum. it's just a giant unusable disaster - same as all unmix algorithms before it. nothing groundbreaking here unfortunately, and now i spent 300 bucks on this useless software because there was no trial... feels like that was probably by design. if you're looking for a way to separate stems from a song, this aint it.
www.steinberg.net/spectralayers/trial/