Split Melody and Vocal Stems to separate related mixer channels

[WillerZ]

Feature Description

For the track loaded to deck 1, I would like to be able to assign the vocals stem to mixer channel 3 while leaving the remainder of the music on mixer channel 1 (and of course similarly for deck 2 and mixer channels 2 and 4).

The mixing problem this would solve for me is when I'm trying to get from one vocal-heavy track to another, e.g. Sugababes Red Dress <-> Girls Aloud Call the Shots. The workflow I want is to split the channels for the both tracks, bring in the new melody, bass-swap, fade out the source vocals, bring in the new vocals, fade out the old melody, then (optionally) un-split the new track back to a single channel.

Some hardware controllers and other software have per-stem volume on the EQ knobs. This isn't enough for my use-case - I need a mixer channel so that I can cue the muted vocals and bring them in when it makes sense.

Related feature #13701 is I think different, because that talks about deck splitting. I don't want the vocals and melody on different decks because I need all the other deck stuff to happen to both the melody and vocals: cues, loops, rolls, slicers, speed changes, key changes, etc should be perfectly synchronized.

This should work well for anyone with a 4-channel mixer and 2 deck controls (almost all 4-deck controllers in other words).

[JoergAtGithub]

You can load selected stems into a sampler deck. Did you tried the load menu in the library context menu?

[WillerZ]

I don't want them on a separate deck of any kind because the transport control needs to remain synchronized across all stems.

[Eve00000]

In a controllermapping you can achieve this by assigning a function to a 'split-vocals-button' which

• clones the track from deck1 to deck3
• mutes stem1, 2 & 3 on deck3
• mutes stem4 on deck1

Indeed once deck 1 & deck 3 are playing in this combination, changing the speed (slider or a little correction with a button) on 1 deck will break the sync.
To achieve what you want the 2 decks need to be able to have synced controls, so while the 'split-vocals-mode' is active all controls for deck 1 must be doubled to deck 3 too (speed, key, effects). This can be achieved in a mapping ... but doing it manually and controlling the volume / effects of each stem is way much more fun but requires appropriate hardware (all stemcontrols for each deck).

Splitting the sound output of 1 deck to mutiple output channels would mean changing the basic architecture of Mixxx, as the output of each deck is created pre-fader: source + speedchanges + keychanges + stemvolume + stemeffects + effects are combined and sent to the assigned outputchannel.

To achieve mixes like you descibe I use this logic of cloning (I use an external mixer and lossless tracks/stemcontainers), the downside is that this logic requires a performing machine as for mixing 2 stemtracks (stemcontainers) you need to play 4 stemtracks = 16 streams + extra speed / key / effects calculations.