Is there a 'one voice' mode in Dorico?

In a way it’s really smart to always consider a note a part of a voice, but it means a few extra things to consider when entering notes, dealing with rests and so on.
However, sometimes it would be easier if one could deal with a ‘voiceless’ mode, where one could enter notes freely without worrying about rests popping up. There are of course many ways to deal with the situations associated with ‘voice mode’ always being on: like chord mode, different colours for different voices and so on.

It seems to me (maybe because I’m used to this over the last 25 years) that two modes is the best solution, so first of all: is there a ‘voiceless’ mode in Dorico already?

A voiceless mode would need some intelligence in Dorico’s hand in terms with sorting out how to deal with rests and other voice related situations. But if we can enter notes in a simple mode, and then have Dorico do it’s best to get things right automatically, we could enter the more advanced mode later, meaning full control over each voice, with all that implies.

I’ve searched this forum re. a related topic (“how do I assign a note to a different voice?”) - but in the threads I found, I could read that this wasn’t implemented yet. If it’s not in there yet, I hope it will come soon, and also that Dorico won’t rely on MIDI channel numbers for it’s polyphonic voice assignments. But it would be great if each node had a voice assignment number next to it (in a separate inspector or info window) which would let us assign any note to any voice we want in a way that would be as simple as changing MIDI channel for a note in an event list.

No, Dorico doesn’t have a “voiceless” mode, and nor, on the face of it, is it something that I think it needs to have. If you’re going to have notes, they have to be in a voice of some description, so even if we were to try to hide that fact from the end user, I’m not really sure what would be gained. In general having notes popping up in funny places without any rests appearing to make sure that bars add up or that notes can be correctly placed in time unambiguously would seem like a pretty bad idea to me.

I’m not understanding what it is you’re looking for with a “voiceless” mode. As far as note input is concerned, as long as you don’t deliberately add a voice, you’re in a default setting without the need to concern yourself with voices.

If you’re wanting poly-rhythm within a measure that would need to be entered via more than one voice, all you have to do is enter it in and then use the option to hide extra rests or use the “starts/ends voice” commands. It seems it would be rather murky waters to allow a notation program to allow a voice to randomly start without requiring the right math within the measure. As it is, Dorico is extremely (and that’s putting it mildly) flexible how they’ve already designed it. Maybe I’m not understanding what you’ve meant by your initial post. If so, sorry!

Hi, by ‘voiceless’ mode I mean a mode which is similar to MoTu’s QuickScribe, Logic Pro’s Interpretation mode (derived from Notator/Notator Logic) or what programs like eg Musicator did on PCs many years ago.

Just to clarify: I’m a piano player, this idea isn’t meant for situations where you write a two voice divisi section for V1s. In such cases, you’ll always think in voices anyway.

If you record a MIDI file in a DAW, and imports that file into Dorico, Dorico already does its best to interpret the material- and probably does that better than most, if not all others, even if no human did anything to manually assign voices to what was played, or manually told Dorico where there’s need for rests, or where the are or are not chords.

When imports a real time MIDI recording as a MIDI file, or enter notes, either with a mouse or with a keyboard - the user will be in a “I don’t know where rests need to be placed, or which voice each of these notes should be assigned to” mode. Dorico will be in a "Dear user, just enter single notes, intervals, chords etc without thinking about voices or rests, and I’ll do my best to display what you enter in an as good way as when I deal with importing MIDI files, MusicXML, or real time MIDI recordings’-mode. So the user doesn’t think of voices; Dorico does.

Call it a sketchpad mode if you will, where all the already existing - and brilliant - code in Dorico is used to interpret what the user enters. Of course: at some point, both D. and the computer will immediately need to think about how all these piano notes shall be displayed, but the user doesn’t need that until he sees how successful D. was at displaying things.

That’s the default mode which score users in Logic and several other apps are used to. There are several ways to automatically and manually define how things shall be displayed in these apps too, including a way to define for each and every note which voice it belongs to. Since Logic unfortunately are using MIDI channel numbers to assign polyphonic voices, it has always been able to easily allow reassignment of polyphonic voices: you just change it’s MIDI channel. But Logic and it’s (lack of) development is frustrating both in the score and other areas.

And even if Logic isn’t at all a dedicated score app, it has nevertheless been used by users to prepare parts for musicians and orchestras for some years now, almost never based on a fully automatic interpretation of what has been played or entered, but based on a combination of the computer’s automatic interpretation and the use’ manually editing the result afterwards.

I can already achieve this, by entering notes manually in a DAW and import the music into a score app. Or I can enter notes, one at a time. But for someone who wants the best of two worlds, it should IMO be possible to achieve what I can achieve with combining eg step input mode in any ‘dumb’ DAW with importing that material into Dorico… without needing that other program.

This could probably be dealt with in at least two possible ways by score apps. In Dorico’s case, it could see the incoming MIDI Notes from a MIDI or computer keyboard (or mouse) the same way it sees a MIDI file or real time MIDI recording when that will be implemented. In other words, it wouldn’t be an actual ‘voiceless’ mode, because Dorico things in terms of voices when it does that. Alternatively, the score app/Dorico could possibly add an abstraction layer called Sketchpad Mode or similar, which will ‘warn’ the user that this isn’t the best Dorico can deliver, and tell users that this mode is only a way to help the user immediately see a result which looks as correct as what he already gets when importing a file or record MIDI notes from a MIDI keyboard.

There are so many situations where we’d need to do some serious thinking in order to define voices when we enter nodes manually, for instance a piano chords which starts a single voice but which during a couple of beats have developed into a chord with, say, 7 sustained notes. But I’d rather use by brain to think of ‘music’ than of ‘voices and rests’, at least in the creative process.

And I’m still a Dorico novice, so maybe there actually are relevant solutions in there which I’m not aware of. But my experience so far is that when I enter a note, defining which voice it shall belong to is a part of that process. And trust me; I hope I’m wrong! :slight_smile: .

Of course I can enter notes and deal with ‘Start Voice’, ‘End voice’ and all that, but I firmly believe that there’s a much simpler - and good - way for users to enter notes into score apps than to thinking about voices. I already have had situations where I see a rest which doesn’t seem to belonging to a voice, and I can deal with that as well - by enabling the colour mode view which tells me which voice each of the notes and rests are assigned to.

But couldn’t a sketch-pad mode where none of this would be needed (at first) potentially develop into becoming a major time saver, even for very complex music (maybe especially for very complex music)?
I think so.

It sounds like what you are describing is what we would call ‘voice detection’ or ‘voice separation’ for MIDI input. A good voice separation algorithm greatly reduces the amount of editing required after importing or recording piano scores. This is something that we would like to do in the future.

Yes, that’s one of the two possible ways o deal with it (that I can think of). When dealing with piano music, one is regularly switching between everything from 1 to 7+ voices, which also can happen inside one single bar.

The other way would be do see all the notes as simply a ‘piano voice’, and add code which would display it as of Dorico already had analysed it the way it analyses MIDI files (or, in the future, detects voices from a MIDI input). That’s what a few apps does already. And Dorico already allows one ‘voice’ to have several polyphonic voices (notes), so maybe a combination of several methods would do the trick. When polyphonic mode is activated in Logic for a piano piece, all notes will usually appear in treble clef (if one uses a ‘score set’ with two voices only, where one voice is assigned to the upper stave (usually treble clef), and the other to the lower clef… so one needs to ether automatically or manually (with a voice separation tool) define which of the polyphonic notes that shall belong to each of the clefs (polyphonic voice). Afterwards, one can have a total of 16 polyphonic voices freely distributed between the upper and lower clef. Each of these voices can have as many polyphonic notes as one desires. And, as with D., rests are created automatically, but the workflow dealing with polyphonic voices is what one usually uses when deling with rests.

The ‘classical’ problem is of course that one easily gets too many rests, and these can to some degree be suppressed. But I think there could be benefits for Dorico users if D. would add an extra display mode, with automatic display of rests, as usual, but where a combination of chord mode and chord mode would be able to show a relatively useful display of a piano track already before manual editing had started - based on only two ‘voices’ (one for the upper, and one for the lower clef). By avoiding the traditional voice thinking, one wouldn’t have to deal with all those extra voices, and of course: it should always be possible to convert any track or bar to the traditional ‘voice’ mode which Dorico uses now. In Logic, you can decide this for each bar… so bar 1 can have a polyphonic voice mode, the other one could be based on the more primitive Interpretation mode and so on. Actually one can have different modes for every beat or even every 1/16th note position if one desires, and the UI for all this is quite simple: on just divides the region in several parts, and each part can have it’s own setting.

Regarding ‘voice separation’ for MIDI input, this somehow has to happen in real time, otherwise one could just fool Dorico’s existing code and take every bar, once it has been entered, and treat it as if it was an incoming MusicXML or MIDI file. But until a bar is completely entered, I wouldn’t mind seeing a semi-perfect ‘voiceless’ (where all notes; both chord notes and single notes belong to only each ‘voice’ pr. clef/stave) display for that bar - with automatic rests a la other music programs’ ‘interpretation mode’, ‘quickscribe mode’ etc… and where things are displayed as correct as possible without using the advanced solution which implies dealing with a lot of rests etc. Sorry for the long sentence.

Don’t get me wrong - I’m not here to promote Logics’ solution, and both QuickScribe, Musicator and others has/had a better automatic display of polyphony than Logic has today. But since the Logic’s score development is more or less dead, and Dorico already is advanced on so many levels, why not also add the kind of functionality that would attract all those potential users which aren’t professional engravers (which there realistically has to be very few of), but who wants a result looking as good as if it has been treated by a pro engraver?

Regarding these DAW users… they (no; we!) usually/often know enough about notation to discover when something is wrong, but want to focus on getting the music right a lot more than wanting to deal with getting the notation right. So I think my suggested ‘idiot mode’ :slight_smile: could be a great bridge between composing and getting a great looking result, without switching into nerd mode in the midst of the composing process. :slight_smile: