Dialing in tones using Moises app

I’m curious how well this app could do if you isolated a guitar track, exported it, and then used the axe iii tone match.
 
Very nice!. Yes, if you search for moises you'll find plenty of discussion of it on the forum.
 
I usually load a song from My Music and it isn't a perfect isolation AND I'm surprised how different guitars sound In the Mix as opposed to isolated. It is a great too for learning guitar and bass parts...but you have to remember I'm old enough to remember wearing out records moving the needle back and forth or rewinding cassettes.
 
I got better results using Izotope RX audio separation. But I haven't find a better solution yet. Sometimes Lalal does a better separation when keyboards are present in the mix.
I need a tool that can separate guitar and keys.
 
It doesn’t always work. Almost always includes organ and other guitars on the same stem.
yeah it leverages a model called "spleeter" I believe

https://github.com/deezer/spleeter

that was trained primarily to split drums, bass, vocals and everything else. Guitar is "everything else". Someone should train a guitar oriented model to get minus-guitar tracks and it would work better. In order to do that you need the final mix and the stems to train the model to recognized the individual parts. Of all places where that dataset can be sourced this forum is probably the best bet.

I have some ML experience and work for a company that's full of people that could help (Google) a lot of musicians that could chip in. Send me a private message if you're interested contributing.

edit: yep, they use spleeter, it's here https://moises.ai/blog/how-to-separate-audio-tracks-online/
 
yeah it leverages a model called "spleeter" I believe

https://github.com/deezer/spleeter

that was trained primarily to split drums, bass, vocals and everything else. Guitar is "everything else". Someone should train a guitar oriented model to get minus-guitar tracks and it would work better. In order to do that you need the final mix and the stems to train the model to recognized the individual parts. Of all places where that dataset can be sourced this forum is probably the best bet.


FYI, "Guitar" is separate from "Other" in Moises. I've noticed Moises has improved significantly since the date of that blog post, but I don't know any details about how it has changed.

moises.png
 
FYI, "Guitar" is separate from "Other" in Moises. I've noticed Moises has improved significantly since the date of that blog post, but I don't know any details about how it has changed.
I don't know how they do it but spleeter most definitely doesn't support extracting guitar alone. That hasn't changed. from https://github.com/deezer/spleeter
Spleeter is Deezer source separation library with pretrained models written in Python and uses Tensorflow. It makes it easy to train source separation model (assuming you have a dataset of isolated sources), and provides already trained state of the art model for performing various flavour of separation :

  • Vocals (singing voice) / accompaniment separation (2 stems)
  • Vocals / drums / bass / other separation (4 stems)
  • Vocals / drums / bass / piano / other separation (5 stems)
and a discussion on training a guitar specific model

https://github.com/deezer/spleeter/issues/3

which brings up the lack of an adequate data source. So... send me your logic projects?
 
Give RipX a try. As someone else mentioned you allways get both synth and guitar together when you unmix stems, but with RipX its (in some cases) quite easy to get rid of the synths with a little manual work. Best unmix software I have tried so far.
 
bumpy bump bump. I finally got around to trying moises after seeing it pop up across various threads for years. Why did i wait this long?! helluva handy app I must say. So many uses for it. Practice tool, making backing tracks to jam over etc. I'll be using this a lot. I even exported a .wav file that I loaded onto my Yamaha MODX with the keys/synth parts removed. Will use equally a lot for guitar playing too. Sorry Eddie, but i'd like to jam with your band a few songs okay bud.
 
Back
Top Bottom