I have a SpectraLayers 8 license. Not sure I know how to do this but I can give things a shot at some point with some effort and/or guidance.Lord Reith wrote: ↑Sat Jun 18, 2022 1:49 am Okay, I can provide the wave file of the melody. Now is there someone out there with deep pockets who has Spectralayers Pro who can actually do the job? I don't use warez so there is no possibility of me doing it myself.
Is there any demixing software like this?
- InnerLight
- Posts: 857
- Joined: Tue Jan 04, 2022 12:32 am
- Has thanked: 39 times
- Been thanked: 5 times
Re: Is there any demixing software like this?
Re: Is there any demixing software like this?
[quote=BDJ post_id=57777 time=1655492817 user_id=375
Sure, this is possible in MAGIX Spectralayers Pro. And I assume in similar packages as well.
You would record the vocal melody from midi into a 'wav' type signal (available in several softwares) and insert it into a song in Spectralayers. You would then select this signal and its Harmonics (available in Spectralayers), cut it, paste it into a separate track, and Bob's your uncle.
[/quote]
The human voice is not perfect. The vocal will waver around a note/frequency unless it's a modern auto-tuned monstrosity. Unless your midi follows exactly every single human imperfection with pin sharp accuracy alongside its timing and with the exact same harmonics, It Isn't going to work particularly well.
Izotope RX pro has the ability to remove things like foldback and headphone spill, but both the source that fed the headphones or foldback needs to be available and in 100% sync for this process to work a fraction of a millisecond out, and it doesn't work so well the more out it is timing wise the wosre the results. - The same principle as above.
Sure, this is possible in MAGIX Spectralayers Pro. And I assume in similar packages as well.
You would record the vocal melody from midi into a 'wav' type signal (available in several softwares) and insert it into a song in Spectralayers. You would then select this signal and its Harmonics (available in Spectralayers), cut it, paste it into a separate track, and Bob's your uncle.
[/quote]
The human voice is not perfect. The vocal will waver around a note/frequency unless it's a modern auto-tuned monstrosity. Unless your midi follows exactly every single human imperfection with pin sharp accuracy alongside its timing and with the exact same harmonics, It Isn't going to work particularly well.
Izotope RX pro has the ability to remove things like foldback and headphone spill, but both the source that fed the headphones or foldback needs to be available and in 100% sync for this process to work a fraction of a millisecond out, and it doesn't work so well the more out it is timing wise the wosre the results. - The same principle as above.
- AllTogetherNow
- Posts: 427
- Joined: Sat Jul 10, 2021 10:25 pm
- Location: Somewhere
- Been thanked: 56 times
Re: Is there any demixing software like this?
The human voice is not perfect. The vocal will waver around a note/frequency unless it's a modern auto-tuned monstrosity. Unless your midi follows exactly every single human imperfection with pin sharp accuracy alongside its timing and with the exact same harmonics, It Isn't going to work particularly well.Bazilnerk wrote: ↑Sat Jun 18, 2022 6:27 pm [quote=BDJ post_id=57777 time=1655492817 user_id=375
Sure, this is possible in MAGIX Spectralayers Pro. And I assume in similar packages as well.
You would record the vocal melody from midi into a 'wav' type signal (available in several softwares) and insert it into a song in Spectralayers. You would then select this signal and its Harmonics (available in Spectralayers), cut it, paste it into a separate track, and Bob's your uncle.
Izotope RX pro has the ability to remove things like foldback and headphone spill, but both the source that fed the headphones or foldback needs to be available and in 100% sync for this process to work a fraction of a millisecond out, and it doesn't work so well the more out it is timing wise the wosre the results. - The same principle as above.
[/quote]
Have to agree.
Thats why I riddicule in my post above
My humble thought,,, but I am an analog man....
Tegards
ATN
Whats in that jar on your shelf? Tar?
No...its liquid Vinyl for making my dream collection.
The one with the Golden lid too?
Thats exclusive for Lord Reiths stuff
No...its liquid Vinyl for making my dream collection.
The one with the Golden lid too?
Thats exclusive for Lord Reiths stuff
- Kwai Chang
- Posts: 1574
- Joined: Tue May 25, 2021 3:46 am
- Has thanked: 331 times
- Been thanked: 193 times
Re: Is there any demixing software like this?
This conversation reminds me of the time Gilligan found the bush with the mind-reading seeds:Bazilnerk wrote: ↑Fri Jun 17, 2022 5:50 pm If humans could sing perfectly, that would be a start, but the imperfections in the voice make that difficult as well as the waveform changing every time the voice changes syllable.
Currently when music is demixed it removes the sound and leave a hole where that sound was, so everything in the same frequency gets separated. Sound, when intertwined with many other things is far too complex for computers to simply separate anything you like. I think it will come, but we aren't there yet.
Professor: It's scientifacally impossible to read another person's mind!
Ginger: Go ahead Gilligan, read my mind for the Professor! Who am I thinking about?
Gilligan: Hmmmm...Rogery Hudpeck!!! Rogery Hudpeck???
Professor: See? It's Impossible!!!
Ginger: No! Gilligan's right...I couldn't make up my mind. I was thinking of Rock Hudson AND Gregory Peck!!!
I agree with Bazilnerk! It will take time, but it's already being refined at an exponential rate. It might be far-out to disassemble Beatles music. But, it's scary to know that machines will be able to sort the din of a crowded room into individual conversations, identify the individuals involved and then plot their locations against recent movement and finally enter the pending concerns that might be urgent or incomplete(warrants, unpaid debt, AWOL, etc). Oh, look!!! Here comes the handcuff-bot! And, ALL of the bots are named HAL, Johnny Cab, Or Judas.
KC
- Lord Reith
- Posts: 4711
- Joined: Thu Feb 18, 2021 8:22 am
- Location: BBC House
- Has thanked: 148 times
- Been thanked: 4090 times
Re: Is there any demixing software like this?
Okay, that's kind of different to what i was thinking. I was thinking more of guiding an AI demixer by giving it the melody of the harmony, so it can decide which vocal is which. SL is a manual spectral demixer isn't it, so is 'dumb" in the sense that it doesn't know what anything is unless you tell it.
Come on, there must be some hackers reading this. Here's your chance to make your name.
Come on, there must be some hackers reading this. Here's your chance to make your name.
- Lord Reith
- Posts: 4711
- Joined: Thu Feb 18, 2021 8:22 am
- Location: BBC House
- Has thanked: 148 times
- Been thanked: 4090 times
Re: Is there any demixing software like this?
No, it was just that I don't think I've ever in my life expected anyone to "respect" me. When I was a kid it was only old people who talked about respect, and now that I'm old it's only teenagers who talk about it. Go figure!AllTogetherNow wrote: ↑Sat Jun 18, 2022 5:41 pm Are U adressing me Sir
Remember I wrote the statement
My best regards
ATN
Women there don't treat you mean, in Abilene
Re: Is there any demixing software like this?
Annotated source separation is a thing but recent research has focused on blind separation.
- Lord Reith
- Posts: 4711
- Joined: Thu Feb 18, 2021 8:22 am
- Location: BBC House
- Has thanked: 148 times
- Been thanked: 4090 times