Keynote detection VST

[robinw]

Bonjour to everyone,

First of all, I know that there was a similar topic back in 2009, but without much useful information for me. Meanwhile, technology continued to evolve and I wanted to ask, if somebody knows of a VST plugin or standalone software that is able to accurately detect the keynote of a short sample loop. I need this, because I am creating a sample pack that consists of chopped and mashed up melody loops. It will be more comfortable for the end user, to have a keynote information. Because I have no midi file and the source material is completely altered, I cant define the keynote by myself.

Salutations!

[stanlea]

https://www.zplane.de/index.php?page=description-tonart

[robinw]

Salut,

thank you for your reply. I downloaded the demo, but this software can not handle short loops. Also it seems, that you have to make out a licensing price with the company, if you want to buy it. I am a bit irritated with that.

[JumpingJackFlash]

this software can not handle short loops.

If you're dealing with short melodic loops, then nothing is going to be able to determine the key with any reliability. In fact, the term "key" in itself might well not even be appropriate in such a context.

Say your melody loop contains the notes A, G, D and C as a simplistic example. What key is that? - The question is meaningless because it depends on how it is used in context. Those notes are diatonic to many keys, and one or more could even be chromatic and yet still used in the context of an overall key.... Not to mention the fact that not all music is tonal (in a key) anyway; there's also modes for example.

Even longer melodies can be treated in several different keys based on how they are harmonised. The chorale settings by J.S. Bach are a good example; in many cases he took the same existing melody but set it in completely different ways, in different keys (and modes).

Determining the key of a short melodic fragment is a bit like determining the rhyme scheme of a poem from just a single line. It's not possible because you don't have enough information.

Remember; music is an art, not an exact science.

[robinw]

Thanks for the advice. I've really seen this topic as a science. I was wondering about the following: If you buy a melody sample-pack, you often find the keynote mentioned in the filename of the samples. How do they determine them in such short loops then? Like I said before, it would be a comfortable thing in my own creations. But if there is really no exact way to do so, I will just leave it. I don't wan't to do things by halves.

[jancivil]

... 'this topic' as a science? Why? What is 'this topic'? You say you saw this talked about before but the information wasn't useful.

'This' cannot be a science. Let me demonstrate: Take four notes that are pretty vanilla, let's go with G, A, C & D.
What can a musician do with these? All kinds of things. It could be something one has the notion to do over a C major chord; over an A minor chord; as a lead in to a D which is the central note. In the latter case it may be a selection of tones from a seven-note mode, or it may be pentatonic. It could fit a number of pentatonics based on G. G could be the central note, any of the four could be, really.

It may be a choice over a C minor chord, or a Bb major chord, or an Eb major chord. The first may be a choice out of C Dorian; the second may be in Bb major or it may be in another key which contains that harmony; the third may be an idea as per Eb Lydian mode. Not one of these ideas is enough to determine key, and in the case of modes, defining key is a red herring. IE: The number of flats for C Dorian is two, which is the key signature for Bb major or G minor, neither of which defines key for C Dorian and absent a certain understanding will be just misleading; modes do not have keys like that.

I've only just begun to come up with things off the top of my head, and none of these is particularly a stretch or peculiar. As JJF said, determining key requires context. You can't simply lasso some notes and define key outside of context.

I was tempted right away to reply, 'you want to ask a musician about this notion'. I don't know if this was useful information to you, but if you want to make things [that are] useful for music you might want to do some more investigation into this sort of thing.

[jancivil]

If you buy a melody sample-pack, you often find the keynote mentioned in the filename of the samples. How do they determine them in such short loops then?

Either there is a solid way to tell, in context, presented within an actual idea which clearly does belong to key (perhaps a musical passage complete enough to draw the conclusion), or they are cheating.

[Sendy]

One day I'll write a track called "Drone in C". It'll be an evolving soundscape played on a mid G. (With a fractional, subliminal blast of C at the end to resolve it )

[jancivil]

"drone on [in] G" and "melodic loops" is two different things though. I think my statements hold. There could well be a 'melodic loop' that is conclusively in key (a cadential ornament in the baroque style, eg) in very little time, but easily believing this is a science (esp. given the purview of 'loop packs') is another matter.
But the cadential ornament on G can work on G in the key of C, so it is difficult to even construct this truism.

[Sendy]

Eh, it was just a (probably rather poor) joke about the fact that the less context you have about a bunch of notes, the more ambiguous it's key or harmonic interpretation is. Reductio ad absurdum.

I should probably stick to engineering jokes

[jancivil]

I understand reductio ad absurdum. It looked like it served to say the opposite. I don't know if assigning 'G' as a key to a long drone on G is so bad necessarily.
I think my objection to the concept is larger, that people see a thing tagged with key as glibly as this suggests is a sign of the incurious.

I try to find the hole in my argument before I send it. I think as a truism, it turns on harmonic function; does it contain that or the like.

[Sendy]

It was supposed to be a drone on G, called "drone in C", the joke being that smaller and smaller amounts of notes become more and more ambiguous, with a single note being able to harmonize with pretty much anything. It's that ambiguity that makes music interesting. Boards of Canada are really good at doing this, they talk of chord sequences appearing "from the diagonal" which is a nice way of visualizing it.

[jancivil]

Ok yeah, 'drone based on the tone 'G'' is not through itself enough to state 'key of G', which technically states 'G major' or 'G minor'.
Concretely there is a lot of drone-based or ambient where the center does not so much move about. However I have ambient things that may as well be 'atonal'.

Typically a raga is going to have a real obvious central tone, 'Sa'. But there is Marwa raga where Sa really does not happen to be the tonic, and the basic drone note if a tamboura is used is a minor third down and the P5 from Sa is excluded. Talk about ambiguous. The mood is 'ominous dusk', and it really is achieved through this device.

[ccDuckett]

It's reading threads like these that remind me: having a decent ear when you're younger is no excuse to avoid learning (which is exactly what
I did, back in the gangly angry years)... having a slight knack goes a short way in the long term.
Kids, do not cripple yourself musically through the magic of "I don't need to learn this boring-a** bullsh** "..
(Looking at young self: )

[robinw]

Hey, first at all I have to admit that I really was a bit lazy with learning musical theory. I'm no kid anymore, but in fact I was always an intuitive musician and I went very well with that till today. But that doesn't mean that I would not want to take a closer look on it, what is also the reason why we are having this conversation here. "Threads like this" like you say are helpful to me, because I now learned that the keynote detection has to do a lot with interpretation, what I did not know before. A complex thing, and I think I will do some more research in that.