A combination of triggers and notes

I have been using Dubler a bit now and sadly it doesn’t really do what I hoped it would do.
So far it has been mostly useful for creating rhythm, but I think that with a few tweaks this could be really useful and my idea builds on what the product already does.

I want a combination of triggers and notes (controls). it should work like this

  1. Designate a trigger like “La” to begin note (just like the trigger does now). The note should be created with the pitch I sing at the time the trigger is triggered
  2. when I stop singing the note should end
  3. There should be an option (this is the important part) that restricts note creation outside a trigger, so even if I change the note/pitch nothing should happen.
  4. singing a tune with “la la la la la” should give me the correct rhythm and 5 notes with correct pauses in between.

In essence this means that there is only one note pr. trigger - right now, even with key restrictions - I get a lot of notes and they’re totally out of sync with the rhythm I was singing.

What I am proposing will allow the average user (all the not great singers) to accurately sing a tune in a rhythm and while it might not be the correct note that is always triggered (even with key restrictions) there will be much less for me to edit afterwards since the rhythm is correct and there is only one note pr. trigger.

This would be hugely useful and to be honest it was what I thought the Dubler would do when I backed the project.

Regards Rasmus

2 Likes

I agree, this would be immensely useful.

Hi Rasmus,

Love this - you’ve eloquently summed up a conversation that we’ve had in our offices a few times - including just before release. What you’re suggesting is possible and we were just starting to prototype it before release but then, when we started to dig into it, we realised it’s slightly more subtle than we thought and not something to rush. So what’s the complexity? Well there are 2 main reasons Dubler doesn’t work this way at the moment:
[1] Pitch accuracy. If you make a 'Laaaa" sound - the ‘L’ part is noisy and without any discernible pitch. It is the “aaaa” part that is pitched. So if you trigger a note as soon as you hear a sound - you will likely trigger the wrong note. So we wait until we know there is a pitched sound. There’s actually a bit more going on than that but this at least presents the challenge.
[2] A lot of users don’t want to have to make a new attack sound to change note. E.g. using Whitney Houston “I will always love you” as a reference. They want to be able to do one continuous stream with many different notes 'IIIIIIeeeeeiiiiiiiieeeeeiiiiii" not “Diii Deee Diiii Deeee Diii”.

To cater for the range of use cases with [2] we created pitch stickiness - a high pitch stickiness makes it harder to change note. Actually we have a prototyped version where a full pitch stickiness makes it impossible to change note from the initial pitch - this is getting someway closer to what you’re suggesting.

Anyway the short answer is - expect this to be something we expand on. Whether that ends up as different modes or more differentiation on the pitch stickiness slider - we’ll see. But we have ideas in the works! In the mean time - to help achieve stuff that’s more rhythmically complex - what i sometimes do is set up a single trigger with the trigger sensitivity down low. Send that to the same MIDI channel as my pitched sounds and then use those trigger notes as rhythmic markers to align other notes to.

Thanks for the suggestions keep em coming!!
George

1 Like

Hi George
Thanks for the reply.

  1. is that really a problem? Since the idea is to limit the number of notes to one per trigger you can let the note start at the trigger point and then “later on” decide what the what is when you get the first sensible pitch data… You could do this by measure the “normal” length of a trigger, the “LAA” sound and wait until that has passed to detect the pitch data. So if the trigger is activated wait xx ms to collect pitch data. I suppose the trigger is very short. When both trigger and controls are activated now the pitch is out of sync (probably because of the trigger data). This will of course give slightly longer input delay.
  2. The idea is that this should be something you activate - like a third option (like a tick box - combine trigger and control). So if you want to be Whitney H. use the controls as they are now. I reckon that the majority of your users are not Whitney H, though, and will instead try to use Dubler to quickly input ideas.
    The stickyness doesn’t work that well in the sense that I cannot sing anything in a particular rhythm and get a usable result no matter what stickyness I use.

Best Rasmus

Hi Rasmus,

For [1] you could do it like that but here in lies the compromise - at the moment we start a note as soon as we’re confident on the pitch. So if we always wait for a confident pitch anyway then the wait ends up being the same. That said we do actually have a plan to half this time - give us a month or two to work on it. For [2] yes different modes may be the solution - though we like to keep the interface as intuitive as possible so maybe rolling this into the stickiness could work. We’ll keep you posted! All the best,
George