deMIX Pro is pretty damn good at vocal separations! Check this out

[ar surround]

If no one comes along sooner, I may be able to try this closer to the end of the year or the first week of the new year. I’ve been using RX for years for voice work, but never the De-Reverb module. The computer and booth I work in is torn apart at the moment as I get ready to make some updates.

If you’re anxious to try it sooner (and no one else has it) I believe De-Reverb is part of the least expensive version and there’s a demo available. I’m not aware of any free version.

Like I said, my "stone knives and bear skins" de-reverb process worked because of what fell out from one of the DeMiX Pro separations. All of the excessive reverb was in the rear channels. However, the front guitar was mixed off center but won't separate without annoying artifacts. I did manage to come up with something that I like using a few tricks. Here's what the front channels now look like. The song is now a 5.1 with the center channel being a low level mono blending of the fronts to help mask artifacts...a trick that @sjcorne taught me.

 

[M-K]

If you split the track up into the four basic stems (Bass, Drums, Other, Vocals) you will often find that the compression is embedded only in the drum track and you can compensate for that.

It depends. Very often the whole CD is limited and loudness maximized. So over compression is applied to everything. Sad but true.

 

[M-K]

Did I get i right: deMix Pro needs a fast internet line to work? It won't work on an offline computer?

 

[ar surround]

Did I get i right: deMix Pro needs a fast internet line to work? It won't work on an offline computer?

DeMiX Pro 4.1, the current version, works off-line.

 

[M-K]

DeMiX Pro 4.1, the current version, works off-line.

Ah, just saw this "Local Processing and Offline Mode".

But:
"

SYSTEM REQUIREMENTS​

• Mac OS 10.9 and up
• Windows 7 and up
• Minimum RAM requirements 4 GB
• Minimum CPU requirements Core Duo 3GHz
• High-Speed internet required

"
Why do they require a High-Speed internet then?

 

[JonUrban]

Here's the deal. I waited to buy this stuff because I did not want to be sending audio to 'the cloud', I wanted to do the work on my own PC. When they came out with this latest version that allowed local demuxing, I spent the money.

In RX8/RX9/RX10, on my PC, it took about 7-9 minutes to extract a track from the source, like vocals, or bass. Well, DeMUX Pro took about the same time. However, QQ'er MixerRog told me to try the cloud based feature of DeMUX. Being skeptical, I did. And wow. It's so much faster. I happen to have fast internet, but on the whole, what takes 7-9 minutes locally takes about 3-4 in the cloud, including the upload and download times. This is for me with my equipment and my internet.

I think the requirement in the spec's is for downloading the program and updates

 

[ar surround]

Why do they require a High-Speed internet then?

I think the requirement in the spec's is for downloading the program and updates

What Jon said.

 

[M-K]

What Jon said.

Therefore 30MBit/s +- is ok. Although that's not a fast internet line, isn't it?

 

[ar surround]

Therefore 30MBit/s +- is ok. Although that's not a fast internet line, isn't it?

My internet is slow as molasses in comparison...I just tested at 4.74 Mbps...and I have no problems with DeMiX Pro. So you will be fine with yours at more than 6 times what I have. I don't separate on their cloud servers, just locally. I need to upgrade into the 2020's!

 

[Stephen W Tayler]

I'm finding DeMix Pro incredibly useful for cleaning out spill in live multitrack recordings. There is one issue that when processing a mono 96k WAV file in 'local' mode it crashes the program. It does work using the 'cloud' solution though, and I believe they are trying to rectify and update it. It does work in local mode at 44.1k, but it seems to crash at other sample rates when processing mono files..

It is brilliant for so many of my purposes, though!! SWT

 

[ar surround]

I'm finding DeMix Pro incredibly useful for cleaning out spill in live multitrack recordings. There is one issue that when processing a mono 96k WAV file in 'local' mode it crashes the program. It does work using the 'cloud' solution though, and I believe they are trying to rectify and update it. It does work in local mode at 44.1k, but it seems to crash at other sample rates when processing mono files..

It is brilliant for so many of my purposes, though!! SWT

I think that @J. PUPSTER had the same crashing problem with a mono track. Did you try faking it to stereo and running it again?

 

[Stephen W Tayler]

I think that @J. PUPSTER had the same crashing problem with a mono track. Did you try faking it to stereo and running it again?

Yes and it worked , but that misses the point when I use multiple mono tracks that need to be loaded back into ProTools as mono tracks - making them stereo uses more space and complicates the routing!

 

[JediJoker]

I think the requirement in the spec's is for downloading the program and updates

I'd imagine it's for the cloud functionality, actually.

 

[Stephen W Tayler]

When I spoke with support I learned that the 'Cloud' processing is not only faster, but a different algorithm, and actually creates subtly different results - I was encouraged to try both 'Local' and 'Cloud' to see which worked/sounded best, depending on the material and the purpose of the separation.

 

[J. PUPSTER]

Well, hopefully they get that crash event "deBugged Pro" soon. Would rather put my time into working the music rather than having to rig some work around.

I just ran a crash on a mono track, and then viewed the Win. 10 report in the Event Viewer, but didn't see anything that I could decipher (but of course I'm no Computer Tech guy.) I was thinking it could be some kind of .dll conflict with another program, but IDK.

As good as the program is, I can see lots of potential for improvement in the coming years

 

[ar surround]

When I spoke with support I learned that the 'Cloud' processing is not only faster, but a different algorithm, and actually creates subtly different results - I was encouraged to try both 'Local' and 'Cloud' to see which worked/sounded best, depending on the material and the purpose of the separation.

I got slightly different results when I tried a test of "local" vs. "cloud" processing. However, I have also sometimes gotten different results when running "local" processing on the same file.

 

[Stephen W Tayler]

I got slightly different results when I tried a test of "local" vs. "cloud" processing. However, I have also sometimes gotten different results when running "local" processing on the same file.

I also occasionally try and process the original audio before importing it - using level, EQ, compression and various forms of enhancement, I found it can help to distinguish certain elements for the algorithms to detect and separate. This can often make a big difference.

 

[ar surround]

I also occasionally try and process the original audio before importing it - using level, EQ, compression and various forms of enhancement, I found it can help to distinguish certain elements for the algorithms to detect and separate. This can often make a big difference.

Do you have and example, Stephen? I know that a lot of this is a wing-it-as-you-go sort of thing.

When it comes to separating lead and backing vocals, I have found that creating a file that repeats a difficult vocal passage over and over and over often yields at least one iteration that is better than a simple straight processing of the entire song.

 

[Stephen W Tayler]

Do you have and example, Stephen? I know that a lot of this is a wing-it-as-you-go sort of thing.

When it comes to separating lead and backing vocals, I have found that creating a file that repeats a difficult vocal passage over and over and over often yields at least one iteration that is better than a simple straight processing of the entire song.

No - sorry - I can't share what I am working on, most of my mix projects are covered by NDAs!! And I'm not allowed to talk about anything until official announcements from the labels.

 

[ar surround]

No - sorry - I can't share what I am working on, most of my mix projects are covered by NDAs!! And I'm not allowed to talk about anything until official announcements from the labels.

That's funny, because just before I read your post, I was flipping through the Azure d'or booklet looking at the credits. And there was your name! I won't ask you any more detailed questions as I understand your professional position. Also, as strictly a hobbiest/hacker, I probably will be clueless as to what you are saying anyway.