How to... Harmonizing a melody? (starter level)

Chords, scales, harmony, melody, etc.
RELATED
PRODUCTS

Post

I would have thought a simple answer might be:

To find the 'key' of a song, press a note on a keyboard while the track is playing, change the note until it 'sounds right' for the most part of the track. One should 'sit' in place. That's the key you're in.

Harmony - even if you know nothing about musical theory, for each individual note in your main melody, isolate it temporarily (loop or something), then play different notes on the keyboard until you hear a note that isn't the same as the melody root note, but that sounds good, and compliments it? Simple?! :) Repeat for all notes in melody.

As someone else said, start with notes, rather than chords.

Post

SFA wrote: BTW, I wouldn't even know where to begin trying to create a harmony for a melody. Write your progression first, then write a melody over it.
This is very good advice. And it's most likely what songwriters have been doing since all the times. Really, just like learning some open string chords on the guitar - and all of a sudden you're in writing "where have all the flowers gone" land (ahem, ok... great example ;)).

But really, you gotta get used to the sounds of chords with some melodies on top inside a key. After a while these become very natural - they should actually already sound very familiar before you start with, just that after a while you know what's happening "behind the scenes".
This is when you may want to start with re-harmonisation. Or other things such as modulation.

All this certainly doesn't come overnight, but on the other hand, it's no rocket science either and doesn't even require much (if any) talent. Harmonical basics are pretty much like lower grade mathematics, from their difficulty.
And the best thing being: You can more or less start right away. You don't need to be a master before you start. There's also no real "wrong" results

Oh, some rough estimations:

- All the required "rules" for more or less plain scale-chord-relationship stuff should fit on, say, around 3-5 "remainding" sheets of paper. If at all. Once you got the basic principles of how to build chords inside scales, you can probably even squeeze the essence onto one sheet. Can't be that much to learn then, can it?

- Then there's a few "voicing rules" which you may want to put to some use. These fit on one sheet of paper (I know this exactly, because I have them written down, will probably post them later on in this thread).

- Then there's slightly more advanced stuff which actually requires some more practice as well. Such as putting voicing rules to good use. This requires experience, there's no way around it. Fortunately, you don't need to be a piano or guitar wizard anymore, though. The piano roll is your best friend! Don't get me wrong, I absolutely recommend learning the most common chord voicings on an instrument, but you don't need to be able to speed things up a lot, accuracy is not exactly must anymore either. If you start really slowly, these things can still be sort of managed ("sort of" because you will never finish) in a matter of a few weeks. This still hasn't got much to do with "talent" but more with memorizing skills.

- Then, as the last basic to learn, there's harmonisation. This is when it gets a bit more difficult. As you can clearly see from the different replies and the different approaches towards the melody in the first post.
But still, there's a few basics of a) harmonisation and b) modulation which aren't all that tough to get through.
To harmonize a given melody, several approaches can, will and probably should be mixed. Personally, I'm quite a fan of having the melody play and try out various basslines below it. Having an idea of what key a melody is in certainly helps to find out the "better" bass notes, but the occasional "wrong" note might not hurt either.
When done with the bassline, all that's left to do is to fill the space between bass and melody with chord notes. Which, once you are more or less familiar with your chords inside keys, isn't all that much of a big deal.
Another harmonisation technique would be to just fool around with the most common chords and progressions from a given key.
Yeah, that's put very simple, and it defenitely requires some discipline.
Modulation becomes quite harder. But once you know a few ways to modulate, it's not all that tough anymore. Oh, hold on, it is... but starting with it isn't.
For all these things there's no estimations. They will take a lifetme, nothing less. But once familiar with the basics, you won't notice, because that's when the fun starts.
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

If harmonising were only as simple as in the cartoons:

(especially watch 2:48 until 4:15)
We are the KVR collective. Resistance is futile. You will be assimilated. Image
My MusicCalc is served over https!!

Post

Panda wrote:Get this and study it:

http://www.playjazz.com/AC019.html

Take it as a christmas present or whatever. Have fun,

Nicolaas
To repeat myself: this is a really good course if you want to learn about harmony. As some people already said, harmonic theory is imo not a subject you best learn about on an internet forum (not to say that your efforts are useless sascha, those were sound descriptions), neither is it something you can apply like 'how can i route my midi data there and there'-kind of questions, just like that in a minute. If you want to learn this, you're gonna have to take your time and study and practice, or take even more time and find out by trial and error.

And sorry if i was a bit annoying repeating the christmas-thing; i just thought you were a bit rude to sascha who was trying to explain...

All the best,

Nicolaas

Post

Panda wrote:As some people already said, harmonic theory is imo not a subject you best learn about on an internet forum
I beg to disagree.
IMO, if done right, an internet forum is *perfect* for these kind of things as it's interactive, whereas no book will ever be.
Try to ask any famous book author a question about something you may not understand - a hopeless thing. Try to send him a MIDI or audio file - dito.

Really, if people were able to show a little discipline, an internet forum is the next best thing after real lessons. And perhaps for some things it's even better because you can hear opinions of more than just one teacher.
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

Sascha, I think Panda's implied contrast was face-to-face sessions instead of books. Discipline is relatively rare, too. But yes, assuming everyone involved practices discipline good things can happen with regular success.

Post

Compyfox wrote:, I just went by ear.
And don't stop that.

Rule 1.
Hearing comes first. Theory is pointless without a (good) sense of relative pitch. Don't buy a theory book (well do that as well but...), get a good ear-training teacher or buy a ear-training software/course/book. Ear training can be very frustating at first but it is very rewarding and much more important than any theory lesson.

Rule 2.
Theory is based on what people play. You have seen people writing about "thirds". Well the third was actually DISCOVERED a few centuries ago. That does not mean it was not played before that.

Do you see what I mean? They discovered a pattern. Play a G7 chord an then a C chord. You have heard this little "progression" many many times. Well at some point somebody noticed that; Hey here is a pattern! Let's name it somehow! You can call C-G7 a V-I ending, a perfect cadence, an autenthic cadence or you if will "my blue socks", "the nice ending", the boring cliché ending" or whatever you like. The point is if you can't hear it (on the radio or in a band etc) theory is nothing but names and empty "systems".

What I belive theory does for you is that it is easier to organize music in your head. Give abstract things/sounds names. One other thing theory does for you is that it helps you talking about music with others. It is easier to say G7 or "go up a fifth" than play that "interesting chord" or go up " a bit".

Bosse
www.bojomusic.se

Post

Bosse has a strong point. Most theory things were made up *after* people already played that stuff. Actually, this is valid for all theory, I think.
Yet, IMO it can still help quite a bit.
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

Sascha Franck wrote:
Panda wrote:As some people already said, harmonic theory is imo not a subject you best learn about on an internet forum
I beg to disagree.
IMO, if done right, an internet forum is *perfect* for these kind of things as it's interactive, whereas no book will ever be.
Try to ask any famous book author a question about something you may not understand - a hopeless thing. Try to send him a MIDI or audio file - dito.

Really, if people were able to show a little discipline, an internet forum is the next best thing after real lessons. And perhaps for some things it's even better because you can hear opinions of more than just one teacher.
I see what you mean, and i would totally agree if everything people write on forums like this would be right, but unfortunately, that's far from reality, and so i have found myself several times spending more time checking on information instead of actually learning.
Anyway, this course i was talking about has indeed been taught to me by a teacher, so i guess you're right about the interactivity aspect... Although it also depends on who ou are, i learn better from books...

Another thing is, i was arguing against the assumption that you 'know' something afer someone on a forum has answered a question about it. I mean, for compy to learn this stuff, you'd have to be answering questions for months or years, a 5-page thread just won't do it... I don't say it won't help him, but to really learn it it'd require some self-study and that's where a book comes in handy.

Post

Panda wrote: I mean, for compy to learn this stuff, you'd have to be answering questions for months or years, a 5-page thread just won't do it... I don't say it won't help him, but to really learn it it'd require some self-study and that's where a book comes in handy.
I agree in that some stuff wouldn't be cover-able here.
But, hmmm - I really hate to pat myself on the back too much - I actually learned that stuff. I also learned to teach that stuff. I mean, I have a diploma and all. My final diploms's work, while being guitar centered, covered a lot of this stuff and I've got nothing but very positive comments about it from all sorts of people (I'm actually going to make a book out of it next year).
And (again, I hate that self praising...) I have not found much resources presenting certain things in such a condensed, yet understandable manner.
No, I'm not saying I re-defined the methods of teaching harmonic theory, but over the years I've been reading a shitload of books on the matter, and all of them are trying to make a truly big fuzz about some things.

One example: You will have a hard time finding anybody describing scales as the same thing as chords. This has always been completely beyond me. Scales and chords, while used somewhat differently (really, only "somewhat") are one and the same thing. Both are making use of a given number of notes in a given "parental" scale/key. To me, it'd make absolute sense if chords were named like scales/modes on lead sheets. An Edor (dorian) would be the perfect description for a chords "condition". In lead sheets, you will however usually find Emin7, being asked to interprete it more or less freely.
Now, I don't want to redefine chord notation or anything stupid as that.
But in my "method" chords and scales are indeed presented as an entity - ok, at least more or less. Saves time and there's no conceptual breaks, so the interconnection between the two can be seen easily.
Of course, all the books you find will be called "Chord-Scale theory" or whatever, yet, there's a big fuzz being made about bith of them, often leading to the perception they might be something different, whereas they are *exactly* the same.

Another example: I haven't found a single book presenting the "most wanted" voicing rules (those mainly used in a typical 4-part voicing context) on a single sheet of paper. Some piano books seem to present them fairly well when it comes to typical closed lefthand compings, but they are valid in any context, read: They also apply to guitar voicings, to horn sections, to certain string stuff, to choires, etc.
Of course, they are in no way absolute, they aren't covering *every* situation either, but they are covering 95% of our every day needs when it comes to modifying a voicing.
I will gladly post them, if you are interested (didn't do so yet because I didn't want to cause more confusion, but it's probably too late now anyways). I can already tell you that you won't find them presented like that anywhere else. Yeah, that sounds like a massive self-praise, but hey, I haven't been spending years on the subject for nothing...


However, if Compyfox would try to follow my "instructions" from page 2 (or so), he'd be up and running with the basic stuff in almost no time. At least that's what I know from my personal teaching experience.
Even without any keyboard skills, programming all the given examples in a piano roll wouldn't take any longer than an hour. And it might have a "clicking" effect to actually hear things such as the difference in chords "jumps" and proper voice leading. The same might be true for hearing some of the most standard progressions ever. "Ah, this is what 95% of pop tunes are made of" - something like that.
In addition, I already offered answering whatever questions. And I also try to post things in a way that you could just copy them into whatever document, the only problem perhaps being language barriers. Otherwise you could print them and stick them next to your monitor or wherever you're trying out things.

I know, all this sounds like a massive self-praise, but if I weren't convinced with my ways of approaching things, would I even consider writing a book about that shit?
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

@Sascha:

I would be very interested in learning about voicing rules, especially since you're a pro... :) Maybe you could open a new thread for that topic ?!

Concerning your scale/chord comparison: That's why I like to recommend Frank Haunschild's "Die neue Harmonilehre" (also available in English). It says exactly the same: scales are chords-chords are scales.
Last edited by Gustar on Wed Dec 20, 2006 2:26 pm, edited 1 time in total.

Post

I will gladly post them, if you are interested (didn't do so yet because I didn't want to cause more confusion, but it's probably too late now anyways). I can already tell you that you won't find them presented like that anywhere else. Yeah, that sounds like a massive self-praise, but hey, I haven't been spending years on the subject for nothing...
I'd be interested! Let's see them!

And: you're one of the people here that is definateley allowed some self-praise man, don't worry.

I think I see what you mean in what you're explaining here, and i think i can explain some of the trouble: a book about this kind of stuff is often written by a specialist, a guy who's been spending years to compile his info and so on and will thus be wanting to be complete, scientific. I think that's why most authors don't allow themselves giving the general basic principles of things, but want to build a framework where they can fit everything in. Much as when you study languages in university, where they teach you a load of grammar rules and evolution of literary structures and so on but after four years you can't speak the language, while someone else who went to the country where they speak that specific language will learn it much faster. This can be very frustrating, but if you have the backgrund first and then fill up the gaps, you'll get way beyond what you can learn "on the ground". So i guess it depends on what you want to achieve...

Anyways, I'm saying this because i learned this stuff before i started contemplating the educational methods applied (although i don't have my diploma just yet), and i'm happy i did, but maybe i should be more open for other ways of teaching...

Ah well, just a thought (but i'd still read a book!)...

Nicolaas

Post

Ok, Gustar and Panda, I started a new thread here:
http://www.kvraudio.com/forum/viewtopic ... highlight=
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

One thing to keep in mind is that learning only these dry theory information can be very...er..dry. It's best to couple it with real life examples from songs you like--that way you can see how other musicians you respect has utilized that same theory knowledge. I suggest you go out and get some sheet music from bands and artists you like--they don't have to be notation, just simple tablature with the melody and chords identified. Try playing some of those songs and get a FEEL for how other musicians approached chord progression, melody..etc.

I think theory is only useful if you have the "feel" for composing and arrangement. I've heard a lot of guys who are very knowledgeable about theory, and incorporates tons of advanced theory into their music, but none of it is listenable, because they contained no soul, no feeling, no emotions..etc. Music, much like writing, is instinctual. You can know all the grammar, have a vast vocabulary, but if you don't have anything to say as a writer, you won't be able to write a compelling story.

Post

BosseJo wrote:Rule 1.
Hearing comes first. Theory is pointless without a (good) sense of relative pitch. Don't buy a theory book (well do that as well but...), get a good ear-training teacher or buy a ear-training software/course/book. Ear training can be very frustating at first but it is very rewarding and much more important than any theory lesson.

Rule 2.
Theory is based on what people play. You have seen people writing about "thirds". Well the third was actually DISCOVERED a few centuries ago. That does not mean it was not played before that.

Do you see what I mean? They discovered a pattern. Play a G7 chord an then a C chord. You have heard this little "progression" many many times. Well at some point somebody noticed that; Hey here is a pattern! Let's name it somehow! You can call C-G7 a V-I ending, a perfect cadence, an autenthic cadence or you if will "my blue socks", "the nice ending", the boring cliché ending" or whatever you like. The point is if you can't hear it (on the radio or in a band etc) theory is nothing but names and empty "systems".
Yes yes yes. Listening comes first, and best is listening in front of your inner ear.

I know all the theory mentioned in this thread and quite some more but nevertheless all I write are notes. Maybe I analyze them afterwards, most time not. And by this I can write in about any style from late Mozart to contemporary. It is the ability to manifest an acoustic clue into finger movements, written notes or MIDI files that we need - the more direct the better.

It is a pity that you did not follow my thirds advice, it was a practical one. You would have already found some nice harmonies to your tune that are not easy to explain in first grade tonal theory.


Hannes
Maybe doing jokes without the use of smilies ...

Post Reply

Return to “Music Theory”