Become a Site Supporter and Never see Ads again!

Author Topic: A.I. audio stem software  (Read 3818 times)

0 Members and 1 Guest are viewing this topic.

Offline AbbyTaper

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 230
Re: A.I. audio stem software
« Reply #15 on: December 18, 2024, 04:23:10 PM »
Hope they have a new year sale or something on Spectralayers. That unmix crowd feature seems golden for audience mumbling chatter.

I tried it once, and would describe the results as "mixed".

Offline robgronotte

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 370
Re: A.I. audio stem software
« Reply #16 on: December 18, 2024, 10:55:21 PM »
Hope they have a new year sale or something on Spectralayers. That unmix crowd feature seems golden for audience mumbling chatter.

I tried it once, and would describe the results as "mixed".

I have found the whole program very difficult to use. Very disappointing, because the features seem great.

Offline lavaux

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 3
Re: A.I. audio stem software
« Reply #17 on: December 23, 2024, 09:10:31 PM »
Based on quick research, seems like SpectralLayers was built off of Spleeter, which is the old method of extracting elements. They now use Demucs, which you get for free with UVR (but UVR doesn't have the vocal learning feature that Spectralayers has.

I think MVSEP and X-Minus are the most cutting edge, with UVR right behind them. I use UVR to remove audience, then load it up into Izotope RX and spectrally remove all the talking/coughing/non-applause.

Here's a really great research document that showcases all the current updates and movements in the AI audio separation field:
https://docs.google.com/document/d/17fjNvJzj8ZGSer7c7OFe_CNfUKbAxEh_OBv94ZdRG5c/edit?usp=sharing

Offline if_then_else

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 495
Re: A.I. audio stem software
« Reply #18 on: December 24, 2024, 03:40:02 AM »
There are a few other options that haven't been mentioned in this thread yet.

  • Remix from Acon Digital (https://acondigital.com/products/remix) that's also part of Acoustica Premium Edition. I've got the latter and the tool works great for balancing instruments in a mix. But it wouldn't be my top choice for stem separation.
  • Rip X (https://hitnmix.com) which - according to some reviews by sound and mixing engineers - is supposedly the most advanced tool of them all. That said, the DAW itself in which stem separation is integrated looks like it could need improving.

Offline phil_er_up

  • Trade Count: (9)
  • Taperssection All-Star
  • ****
  • Posts: 1303
Re: A.I. audio stem software
« Reply #19 on: December 24, 2024, 06:59:03 AM »
Have used RX, UVR and SL using many different files and genres. Used the Demus in UVR and found that SL and UVR sounded the best in stem separation and almost similar. RX sometimes has this metallic sound to it. Though found mixed results depending on the genre I used with each software. With RX and UVR they just split stems and that is it. No editing features. SL has these editing features like RX with the addition of the eraser tool, batch processing, select similar and ability to add layers or extract from layers and add to another layer or create a new layer. This is a big step IMO.

Have a fast PC - time to process 70 minute set in following software with 4 stem extraction:
RX - 8 minutes
UVR - 8 minutes
SL 60-70 minutes - you can get around the slow processing time by using the batch processing tool and run it over night.

Tried this over the weekend took a audience recording with crowd noise in the back ground and ran "unmix crowd noise" in SL. Was surprised  how well it took out the chatter from the floor noise. What was left after extracting the crowd noise almost sounded like a SBD though some vocals were taken out with the crowd noise. Then selected the vocals manually in a short section in the crowd noise layer and  ran "select similar". It ran for hours to try and find all of them. Did an ok job. If you extracted just the vocal then ran "unmix crowd noise" it might make a nice sounding audio file. SL is like photoshop and can do many more features then just splitting the files. Though as robgronotte suggests its not as easy program to use and would take time to figure out how to really use it.

====================================================

Thanks to  lavaux and  if_then_else for their input and comments. It is appreciated.

Will look into the software you guys posted about. Thank you!
Everyday is a gift. Enjoy each one!
Forward motion bring positive results.

Offline robgronotte

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 370
Re: A.I. audio stem software
« Reply #20 on: January 10, 2025, 09:52:25 PM »
Well, I still haven't figured out a lot of what can be done with SpectraLayers 11, but I just tried using the "unmix crowd noise" for the first time with excellent results.

I took a 14 second part of a recording where there was a lot of crowd chatter under the main vocals and ran it through UVR5.  Took the vocal file from there and ran the unmix crowd noise in SL11, and it gave me a pretty much perfect separation between the singer and the background chatter.  It also was very quick doing that, although again it was only 14 seconds.   Then I exported the "Foreground" layer, and mixed it back with the instrumental audio file from UVR5.

I'm hoping I can use this method in the future instead of agonizingly trying to remove chatter from vocal sections using the spectral repair feature of iZotope RX.

Offline phil_er_up

  • Trade Count: (9)
  • Taperssection All-Star
  • ****
  • Posts: 1303
Re: A.I. audio stem software
« Reply #21 on: January 12, 2025, 11:07:08 AM »
Well, I still haven't figured out a lot of what can be done with SpectraLayers 11, but I just tried using the "unmix crowd noise" for the first time with excellent results.

I took a 14 second part of a recording where there was a lot of crowd chatter under the main vocals and ran it through UVR5.  Took the vocal file from there and ran the unmix crowd noise in SL11, and it gave me a pretty much perfect separation between the singer and the background chatter.  It also was very quick doing that, although again it was only 14 seconds.   Then I exported the "Foreground" layer, and mixed it back with the instrumental audio file from UVR5.

I'm hoping I can use this method in the future instead of agonizingly trying to remove chatter from vocal sections using the spectral repair feature of iZotope RX.
Good job figuring out how to use the tools. Cool that you used SL then UVR5 then SL.

That is my thought pattern too. Will have to use the tools together to get what you want. RX and UVR5 are good at separating a file into stems though that is all they do. SL does so much more though you have to use their tools then manually edit the output to separate what you want in the file and (Noise, chatter, instruments) that you want to take out.

Tried "unmix crowd noise" and thought it did a good job. It completely took out chatter between songs. Though in some songs it took out some of the vocals that the program thought was crowd noise. Wish there was a way to only select different sections of the file then run the tool.



 
Everyday is a gift. Enjoy each one!
Forward motion bring positive results.

Offline AbbyTaper

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 230
Re: A.I. audio stem software
« Reply #22 on: February 01, 2025, 04:37:34 PM »
Has anyone tried Moises?  Not free, but under $3 per month on an annual contract.

Offline robgronotte

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 370
Re: A.I. audio stem software
« Reply #23 on: February 01, 2025, 07:26:44 PM »
Well, I still haven't figured out a lot of what can be done with SpectraLayers 11, but I just tried using the "unmix crowd noise" for the first time with excellent results.

I took a 14 second part of a recording where there was a lot of crowd chatter under the main vocals and ran it through UVR5.  Took the vocal file from there and ran the unmix crowd noise in SL11, and it gave me a pretty much perfect separation between the singer and the background chatter.  It also was very quick doing that, although again it was only 14 seconds.   Then I exported the "Foreground" layer, and mixed it back with the instrumental audio file from UVR5.

I'm hoping I can use this method in the future instead of agonizingly trying to remove chatter from vocal sections using the spectral repair feature of iZotope RX.
Good job figuring out how to use the tools. Cool that you used SL then UVR5 then SL.

That is my thought pattern too. Will have to use the tools together to get what you want. RX and UVR5 are good at separating a file into stems though that is all they do. SL does so much more though you have to use their tools then manually edit the output to separate what you want in the file and (Noise, chatter, instruments) that you want to take out.

Tried "unmix crowd noise" and thought it did a good job. It completely took out chatter between songs. Though in some songs it took out some of the vocals that the program thought was crowd noise. Wish there was a way to only select different sections of the file then run the tool.

Seems like there should be a way to only select part of a file but I haven't found it. I cut the recording into pieces with another program and then run through just the part I want to use it on. Same thing for UVR5.

Offline bmubart

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 8
Re: A.I. audio stem software
« Reply #24 on: February 04, 2025, 06:15:12 AM »
This is my workflow for noisy crowds (for what it is worth):
- Spectralayers: Unmix Song in Voice + Other (= 2 layers)
- Unmix the Voice further with Unmix Crowd.  You become 2 new layers Foreground + Crowd

You can now re-build the whole track in your DAW (Reaper...) using the layers:
1) Other
2) Foreground
3) Crowd

Listen to the result. 
You can now edit the Crowd-layer on the fly in eg Izotope RX or a wave editor to eliminate crowd noise.
Render
--> same recording with almost no crowd

This is very time consuming but the result is very pleasing (imho).
This doesn't work well when there are singers in the crowd which voice blend with the lead-singer

All unmixing in Spectralayers can be done in batch




« Last Edit: February 04, 2025, 06:26:49 AM by bmubart »

Offline phil_er_up

  • Trade Count: (9)
  • Taperssection All-Star
  • ****
  • Posts: 1303
Re: A.I. audio stem software
« Reply #25 on: February 06, 2025, 06:36:38 AM »
Good work  bmubart. Thanks for sharing.

As TS members figure out how to use these software's  then maybe they could share their processes so others can learn from them.

SL is an advanced software and getting it to do what you want will take practice and time.
Everyday is a gift. Enjoy each one!
Forward motion bring positive results.

Offline bmubart

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 8
Re: A.I. audio stem software
« Reply #26 on: February 06, 2025, 09:28:03 AM »
Good work  bmubart. Thanks for sharing.

As TS members figure out how to use these software's  then maybe they could share their processes so others can learn from them.

SL is an advanced software and getting it to do what you want will take practice and time.

Thanks!

I'm a new user to SL too but it is really not that hard at all for the things I need it: making layers/stems.

I do these steps in batch mode (overnight):
- select the tracks I want to unmix
- select Unmix Song and Seperate them in Layers.  You can choose how many layers (drum, bass, guitar, vocals...) but mostly 'voice' and 'others' will do so each track becomes two wave-files
- select the folder where the batch job has to export the layers (= wave files)

Then: go to the folder with the unmixed layers (= wave file per stem)
- in batch (again):
- choose the voice-layers
- choose 'Unmix Crowd' and select the folder to export the voice layers: Crowd + Foreground

After that you will become all the stems with the voice seperated in two layers: crowd + foreground
Take these stems to rebuild the full track in your DAW:

- others
- foreground
- crowd

(so don't use the original voice-layer)

Play this 3 together in your DAW (= full track WITH crowd-noise)
Eliminate the crowd noise by playing with volume envelopes or by reducing it / deleting it in your wave editor (I use Izotope RX)
Listen again = full track with less crowd noise

Hope this helps?



Offline checht

  • Site Supporter
  • Trade Count: (7)
  • Taperssection Member
  • *
  • Posts: 841
  • Let's meet at alternate foods at the break
Re: A.I. audio stem software
« Reply #27 on: February 06, 2025, 11:15:32 AM »
Anyone else working on a Mac?

I'm having issues with UVR using up all memory if I choose ensemble, then crashing the machine.

Anyone?
MK41s, MK22s; Vanguard V1s matched pair
Schoeps kcy5, nbob actives
Naiant PFA 60v, PFA 48v, IPA
Sound Devices MP-6II; Sony PCM-A10

Recordings at LMA

Offline nulldogmas

  • Trade Count: (6)
  • Taperssection All-Star
  • ****
  • Posts: 1821
    • How I Escaped My Uncertain Fate
Re: A.I. audio stem software
« Reply #28 on: February 06, 2025, 03:54:17 PM »
Anyone else working on a Mac?

I'm having issues with UVR using up all memory if I choose ensemble, then crashing the machine.

Anyone?

I'm on a Mac, but haven't tried ensemble yet. What stems are you choosing?

Offline phil_er_up

  • Trade Count: (9)
  • Taperssection All-Star
  • ****
  • Posts: 1303
Re: A.I. audio stem software
« Reply #29 on: February 08, 2025, 12:22:27 PM »
Anyone else working on a Mac?

I'm having issues with UVR using up all memory if I choose ensemble, then crashing the machine.

Anyone?
Hi Chris,

On old windows PC without a graphics card it would take hours to run UVR with 4 stems and a 2 GB wav file and sometimes would hang my pc.
On new fast win 11 pc with graphics card and 8 gb memory takes only 8 minutes to process 4 stems with 2 GB wav file.

Believe you told me it was only 8 minutes to process one stem on your mini-mac thought that was really fast though was surprised due to you did not have a graphic card.
Have looked at task manager in win 11 when UVR is running and my graphics card is running at 100%. Though it does eat memory too. If creating 4 stems with a 2 GB file you need 4-2gb files output that is 8 Gb in file output from UVR.

So assumed you needed a graphic card for UVR to run fast and not crash your machine.
« Last Edit: February 08, 2025, 12:29:42 PM by phil_er_up »
Everyday is a gift. Enjoy each one!
Forward motion bring positive results.

 

RSS | Mobile
Page created in 0.065 seconds with 40 queries.
© 2002-2025 Taperssection.com
Powered by SMF