PDA

View Full Version : Interval training by song association



sarel_aiber
07-14-2003, 04:44 PM
Okay, so I'm attempting to get to a point where I can hear a melody and just *know* what the intervals between successive tones are.

I've read many resources about intervallic ear-training and although some (such as Ron Gorow's "Hearing and Writing Music") discuss analyzing an interval's qualities (in Gorow's case it's tonality, span and resonance, I think) until identifying it becomes automatic, I don't find this method particularly useful.

I have associated most ascending and descending intervals to songs I know, and I wanted to know if anyone around here has gone through the same process and can atest to the fact that eventually, after enough practice, you stop using the song and just *know* the interval for what it is.

As I said, my goal is to be able to just hear a succession of tones and know definitively which intervals make up that phrase. That is the first achievement I wish to take on the way to playing by ear and transcribing perfectly.

Other tips on those matters are obviously welcome and appreciated :)

Sarel

Bongo Boy
07-23-2003, 05:36 AM
I'm really surprised you got NO response to this post! I have not done this, although of course I do use the Wedding March and Twinkle Twinkle Little Star for P4 and P5 references. Not too sophisticated.

I wouldn't be able to do this with other tunes, unless someone were to tell me first what the intervals were--but it seems like it would be the best way to ear train since musical passages certainly stick in your brain better that the fake tones of an online interval training tool.

Guni at one time was working this topic fairly heavily--perhaps he'll weigh-in with some thoughts.

sarel_aiber
07-23-2003, 10:54 AM
Phew... I was beginning to think no-one was ever gonna reply. Thanks :)

I'm getting the feeling that people at IBM aren't interested in *thorough* eartraining, unfortunately. Oh well, alone in my quest ;)

Maybe someone else will pick on the newlyfound momentum of this thread and join the discussion.


Sarel

WaterGuy
07-23-2003, 01:20 PM
Sarel, no, you are not alone in your quest for a better ear. Ear training has been my personal obsession for the past few months. There were a few threads here at IBM a few months ago, and Guni said he was working on some eartraining stuff then, but since then there hasn't been much on the topic. Unfortunately (hint, hint.)

I have Gorow's book as well, and am of two minds on it. There is some useful stuff in there, but I disagree with some of his basic premises. However, I tend to perfer his descriptions of intervals (i.e. energy and harmony) over association with other songs. Bruce Arnold argues vehemently against such ear training as being too slow and counter-productive for use in improvization. He and others favor a functional approach to relative pitch eartraining, wherein you memorize the sound of each note against a key center. This approach dovetails with the traditional Solfege approach. (BTW - Guni's series on Solfege is a terrific resource (props where props are due, Guni.)) The method of training entails playing a cadance (e.g. I-IV-V-I) to get your ear oriented to the key, and then playing a single note and determining what that note is against the key (e.g. Do, La, Fi or Root, M6, #4 however you prefer to think of it). I've been working at this for a few months and am getting quite good at it, but I'm not convinced that I just haven't learned a neat stupid pet trick. My transcriptions skills are improving, but I still have tons of room for improvement.

loveguitar
07-24-2003, 02:21 AM
talking about this feeling a center in a series of notes, and guessing the note by relating it to the root, I have this question:

I practice relative pitch by guessing the next note relative to the previous, not around a center root. That means if the key is C major and we have A, B and C consecutively, I guess B from A and C from B.

The method you mention is you know C first, and you guess A and B with relative to C. Am I right?

Any IBM advisor can comment on the above two methods?

sarel_aiber
07-24-2003, 09:21 AM
Thanks for all the replies so far.

Waterguy, I don't seem to be able to relate to Gorow's way of identifying intervals, unfortunately. He claims that by doing so you eventually get to a point where you can instantly recognize intervals -- but besides the problems I'm having with recognizing intervals his way, how the heck do you sing an interval from a given note using this method? I'm stumped.

I do, however, like his idea of using 2 methods for the identification of each note -- use whichever method is immediate to identify the note, and then another method to identify it.

He does claim, however, that this becomes immediate and indeed is very practical and fast enough for improvisation. That is the way he does it, anyway, but I haven't heard him improvise or seen his speed of transcription (although I doubt it's slow ;)).

Anyhow, I think it's a long and arduous process before the whole thing becomes automatic, and should probably take a few years of immersion.

I'm currently, as said, in the process of internalizing intervals well enough to use for transcription (some intervals I don't have songs for, so I have to make up my own snippets of melody). Once I'm reliably identifying all intervals up to Maj13 I'll start transcribing, gradually expanding the keys I'm transcribing in, and after a while probably getting back to my sightsinging drilling as well.

I think the critical point is to be able to do everything reliably enough albeit slowly. Once you *know* what's going on, even if it takes too long for practical purposes (such as playing by ear), then constant drilling should bring you to the point where it's fast enough and second-nature.

Naturally, you'll get to the point where you don't need anymore speed (if you mainly play ballads by ear, then playing fast jazz solos is always gonna be hard, unless you do it often enough).

I'd love to keep this thread going, as it gives me a chance to voice things I've wondered about and get other peoples' perspectives and experiences. I find that a lot of IBM is (naturally) guitar-biased, and since I'm not a guitar player the eartraining and other general musicianship stuff is the added value for me. If we can attract more aficionados for the subject it can be very valuable.

Now, my response to loveguitar -- yes, you basically have the idea right. But it's not that you figure out the interval from the keycenter to the note in question. After you drill yourself long enough on scale-degree recognition (you can use the excellent "Functional Ear Trainer" freeware from www.miles.be for that), you basically "feel" which scale degree the note is. So being in C major and feeling the current note is the 6th degree, you know it's A.

Hope this helps, and feel free to ask for clarifications etc.

Best regards,
Sarel

WaterGuy
07-25-2003, 01:49 PM
loveguitar, you’re on the right track with your interpretation, but there’s a subtlety to be aware of. The method is not a matter of singing or thinking of the root (in your example C) and then using an interval distance to identify a note (e.g. thinking “oh, that’s like Twinkle, Twinkle with C, so it must be a G”). It’s a matter of knowing (and memorizing) what each scale degree sounds like. Now, what each scale degree sounds like obviously has a strong relationship to the root, so in that sense, the method can be thought of a relating each note to the root, but not as a distance, rather as a product the harmonic relationship. (There is a great thread going on right now relating to that subjecti don't understand this and it's killing me (http://www.ibreathemusic.com/forums/showthread.php?s=&postid=12858#post12858) .)

An actor/singer named Chris Aruffo has been developing a training program for perfect pitch and documenting his explorations on his website Perfect Pitch Eartraining (http://www.aruffo.com/eartraining/) . His research has inevitably led him to study relative pitch as well, and has some interesting insights into relative pitch. If you read through “Phase 7” of his research, he has some interesting things to say about functional relative pitch ear training and distance relative pitch ear training.

sarel, I agree that this whole ear training thing is a long, arduous process, but will ultimately be very rewarding. I’m happy to have you along for the journey.

perth
07-25-2003, 02:22 PM
my reference is the major scale (solfeggio). i play 21 modes (7 modes of the major scale, harmonic minor scale, and melodic minor scale) twice a day, then hum the 7 modes of the major scale with my instrument, then again without my instrument.

it hasnt made a singer out of me, given me perfect pitch, or given me complex chord recognition, but its certainly helped me recognize not only intervals but different keys and triads as well.

note: ive only been doing it religiously for about 3 weeks.

WaterGuy
07-25-2003, 02:29 PM
perth, cool practice regemin. When you sing the modes, do you use solfege, and if so, do you use "Do" or the relative major degree as your starting point. For example, if you're singing Dorian, do you sing Re Mi Fa Sol La Ti Do, or Do Re Me Fa Sol La Te Do?

Koala
07-25-2003, 04:01 PM
I believe interval recognition does worek best through song association i mean, who couldve given me a better definition of octaves that jimi in the purple haze intro.

perth
07-25-2003, 05:55 PM
Originally posted by WaterGuy
perth, cool practice regemin. When you sing the modes, do you use solfege, and if so, do you use "Do" or the relative major degree as your starting point. For example, if you're singing Dorian, do you sing Re Mi Fa Sol La Ti Do, or Do Re Me Fa Sol La Te Do?
usually i just hum it. sometimes ill say "blah" :). i concentrate on the frequency of my voice rather than what i say. besides, italian scares me :D

sarel_aiber
07-29-2003, 12:17 PM
Waterguy, I agree this whole process, if performed diligently enough, should lead to something truly aweslome -- the ability to subconsciously, without thinking about it most of the time, sight sing, transcribe, and play by ear complex melodies, harmonies and rhythms.

It sounds almost too good to be true, especially when we're stuck here trying to figure out individual intervals ;-)

But the human mind is an amazing place, and once we do it long enough and keep raising the bar as we get better, things should fall into place. That's what I firmly believe.

There are so many open questions as to how it eventually falls into place. In all my searches I've yet to have found someone who started off like us, with a good intuitive yet untrained ear, and whom deliberately trained their ear to achieve the level of performance I've described above.

I truly and honestly hope that in a few years' time we'll be able to be the ones who encourage starting ear-trainers and be able to tell them with conviction that it *does* work and direct them not out of speculation (which is what we do here and is still invaluable to me -- if you don't have someone who's been there, at least have someone intelligent and dedicated along for the ride) but out of personal accounts of success.

Sarel

loveguitar
07-30-2003, 02:41 AM
hi Waterguy and all,

I think I understand what you mean. My query is that:

The way of listening that you and most of the others mentioned
is that you would try to recognize how each scale degree sounds as long as you hear the root.
ie. knowing a C major key, now I play a E, followed by a F. Now you recognize E because you know it's a 3rd from C and you know F because it's a 4th from C. But you don't recognize F from E. You know it's F only from it's relative position from the root.
Lets call this Method 1.

Method 2, which I previously thought was right, was to recoginize E from C beause it's a Maj3rd, and recognize F from E because it's a Min2nd. I don't recognize F from C. Get what I mean?

So my query here is which one is the more correct way of listening? Or both compliment each other?

Method 1 seems more logical and faster in recognizing notes. But it also raises another question: In some solos, it can be so complex that there is no one single key. For eg., some solos improvise the arpeggio of the chords as the progression goes. So it can be playing Bm with B,D and F# and next second it is playing EMaj with E, G# and B. In this sense, do we have a problem with Method 1?

Just some thoughts :)

WaterGuy
07-30-2003, 12:40 PM
loveguitar, you're on the right track.

First, I’d offer a clarification on “method 1” as you describe it. This method can be called “functional” or “absolute relative” pitch ear training. It’s not really a matter of keeping the root of the key in your head and judging the distance to every note. Therefore, I wouldn’t say that “E is a major third from C, and F is a perfect fourth from C.” Rather, I would say “E is the major third in the key of C, and F is the major third in the key of C.” It may appear to be a matter of semantics, but it is an important, if subtle, difference in the way of thinking about it. It’s not a matter of distance from the root, but rather a matter of function within a key.

Now, on to your query… There is a debate as to which is the “correct” approach. Bruce Arnold and his adherents would argue that functional or absolute relative pitch is the best method and ear training based on distances is counter-productive as it will confuse the ear and inhibit the ability to detect function of each note. Ron Gorow and other traditionalists, on the other hand, insist that distance ear training is the only way to go, particularly for the reason you mentioned (i.e. what happens when the key center shifts, or the key is ambiguous.) Gorow goes so far as to say that key signatures are an affectation based on restraints of certain instruments. He argues that ultimately the key is irrelevant, and the only thing that matters for the impact of the melody is the series of successive intervals (i.e. your Method 2 or what I refer to as the “distance” approach).

So which is right? Who knows? As with all things, the truth is probably somewhere in the middle. Alain Benbassat, who wrote the Functional Ear Trainer program Sarel mentioned, says that when he transcribes a song, he will usually start with the distance approach and then check the result with the functional approach. That sounds pretty reasonable to me.

Sarel, I can’t agree more that ear training is the route to better musicianship. I’ve been playing guitar for over ten years, but thanks to tablature, technology and laziness, I never bothered to train my ears. I resolved this year to change that, and my overall musicianship has improved immeasurably in a few short months.

loveguitar
07-30-2003, 02:43 PM
This is getting interesting, Waterguy. Your explanation on Method 1 is very new to me. I begin to see what you mean by the functional method.

You hear an E in a C Major key, your just know it's an E. You don't reference it as a Maj 3rd sound from C or whatsoever. Right?

But that's really hard to understand how to do it and really subtle. It's a bit like absolute pitch, isn't it?

Maybe as I improve my ears, will start to see how all these things actually work :)

WaterGuy
07-30-2003, 03:25 PM
loveguitar, that's right. I don't hear E as a major third from C. Rather I hear E as the major third of C major. I personally use the solfege syllables, so if a song is in the key of C, and an E is played, I hear it as "Mi," which I can then translate to E.

It is a subtle thing, but you can really start to pick up on it. Each note has a personality within the key. For me, Do, Re and Ti were the first ones I could clearly pick out of a lineup, as it were. Do (the root) sounds resolved, stable, like home. Re and Ti sound unstable, but consonant and point like arrows to Do. Those three notes simply announce their presence. Fa is stable and bold, yet pretty. Mi is emotional but happy, whereas Me is emotional but sad. Le is melancholy. Sol is almost as resolved as Do ... and so on.

It's funny that you mention the absolute pitch thing. They are very closely related. And in fact, some people who apparently have some absolute pitch ability, in actuality have very good absolute (or functional) relative pitch and have a good, constant sense of the key of C. That's why those individuals can pick out random notes in isolation, but have great difficulty in musical context.

jazz_cat
08-18-2003, 04:09 PM
This is a really interesting thread!
I share the opinion that both methods (intervallic and functional)
have to be studied. I´ve been doing eartrainging for some years now and came across some interesting approaches.

I think the functional approach should be studied first.

A good reference for the functional method is Steve Prosser´s "Essential eartraining for the contemporary musician"
(http://www.amazon.com/exec/obidos/tg/detail/-/0634006401/qid=1061217313/sr=1-1/ref=sr_1_1/002-8480628-8101620?v=glance&s=books)
It uses "movable do solfege", where "do" is always assigned to the tonic center.
12 syllables are used to sing (and hear) all 12 notes in relation to it.

I studied solfege in a school where the following functional method was used.
It is especially suitable for "absolute beginners" as I was back then.

Major Scale
-----------
1. Arpeggio
First study the arpeggio: sing scale degrees 1(do), 3(mi) and 5(so)
a) within one octave
b) over 2 octaves, jumping randomly from one degree to another
for a "normal" vocal range the possible notes in the key of C would be:
G C E G C E

2. Degrees 2(re) and 6(la)
Now add scale degrees 2 and 6.
Learn and hear th 2nd degree in relation to the root, i.e. if you don´t know where
the 2nd degree is, hear and sing the root first.
Learn and hear th 6nd degree in relation to 5th.

3. Degrees 4(fa) and 7(ti or si)
Learn and hear th 4nd degree in relation to the 3rd
Learn and hear th 7nd degree in relation to the 1st.

The same method was then applied to the

Minor scale:
------------
1. Arpeggio: 1, b3, and 5

2. 2 and b6

3. 4 and maj7 (the harmonic minor scale is studied first)

4. study other minor scales (aeolian b6 and b7, dorian 6 and b7, melodic minor 6 and maj7)


cu jazz_cat

qoody
04-18-2009, 03:36 PM
Many years passed, are there any results on ear training to be shared? Please!

Lorentzyeung
06-04-2009, 11:20 AM
Hi all,

i am curious of you guys progression too, anyone can share?

ragasaraswati
06-14-2009, 12:42 AM
Each note has a personality within the key.


That's right. Also keep in mind the 'personality' of our tuning sytem, the ET, which is to 'autotune' the notes with our brains because, especially in the more cosonant intervals, it will understand the intended just intonation interval and 'hear' it that way. I have listnened to just intonation music and it was more 'passive' no 'autotune' and compromise from my side, no slow beats from its side, and it sounded strangely static. It's because of that.



Do (the root) sounds resolved, stable, like home. Re and Ti sound unstable, but consonant and point like arrows to Do. Those three notes simply announce their presence. Fa is stable and bold, yet pretty. Mi is emotional but happy, whereas Me is emotional but sad. Le is melancholy. Sol is almost as resolved as Do ... and so on.


Cool. Myself I have also developed a 'feel' for all 12 intervals and I challenge any experienced musician to not have done so.

On your remarks.
Are you sure Ti sounds consonant? Also Re I agree it's consonant but I would disagree that it points like an arrow to Do, going from Re to Mi, or even to Sol, sounds as much viable to me. I would argue that as much as Ti leads to Do out of 'necessity' so does from the other side bRe, irrelevantly that you may not use bRe that much. Also I can't seem to make out 'Me', is it bMi?

I would like to add that actually playing Fa over Do you will have a temporal Fa tonality for as long you play it. Do serves as the Sol of Fa in this case, solfegio speaking. Playing Fa under Do, this becomes all the more apparent. To make that bold, Fa is the equivalent of the 'exit' door on a Do room. :D

PS: I used to think that only some intervals are 'emotional' e.g. fifth = open, minor 3rd = searching (yes it can be no sad), fourth = power etc.
But I have come to realize that each and every one of them brings something else to the table. The more dissonant the interval, the finner the 'emotion'. Days I hate the world, I love the tritone for instance. Hating the world is a fine emotion, more coplex than hating someone or a situation. The tritone is the disconnection you may feel in that state spelled out in music.

motown01
06-23-2009, 02:19 AM
Eartraining is so helpful, it just takes a lot of practice.

CrazyTalk
11-03-2009, 04:22 AM
I wanted to know if anyone around here has gone through the same process and can atest to the fact that eventually, after enough practice, you stop using the song and just *know* the interval for what it is.

In my experience, after alot of practice with intervals, it becomes as if you don't use the songs to hear the intervals anymore, but they still remind you of the melodies of songs. So it becomes about the different sensations the intervals have on your ear.

I like to relate it to driving somewhere you haven't been before. The first few times you get lost, but after a while you begin to recognise the 'landmarks', and shortly after that you just know the way by heart, it's not that you don't ever think of the church or McDonalds you have to drive past every time to get to your mums house. Sometimes you think about it some times you dont. Which means remebering the melodies of certain songs is just like, remembering a landmark. After a while you just remember the way there. (To get to Perfect 4th take a left after "Here comes the bride"...)

When I started interval training I tried just by listening to the pitches on their own, however, I found this useless because I had nothing to remember it by. Why create a way to remember something, when someone has already done it? just copy them, but not so much it's plagiarism. :D

boby
04-08-2010, 11:10 PM
loveguitar, you're on the right track.

First, I’d offer a clarification on “method 1” as you describe it. This method can be called “functional” or “absolute relative” pitch ear training. It’s not really a matter of keeping the root of the key in your head and judging the distance to every note. Therefore, I wouldn’t say that “E is a major third from C, and F is a perfect fourth from C.” Rather, I would say “E is the major third in the key of C, and F is the major third in the key of C.” It may appear to be a matter of semantics, but it is an important, if subtle, difference in the way of thinking about it. It’s not a matter of distance from the root, but rather a matter of function within a key.

Now, on to your query… There is a debate as to which is the “correct” approach. Bruce Arnold and his adherents would argue that functional or absolute relative pitch is the best method and ear training based on distances is counter-productive as it will confuse the ear and inhibit the ability to detect function of each note. Ron Gorow and other traditionalists, on the other hand, insist that distance ear training is the only way to go, particularly for the reason you mentioned (i.e. what happens when the key center shifts, or the key is ambiguous.) Gorow goes so far as to say that key signatures are an affectation based on restraints of certain instruments. He argues that ultimately the key is irrelevant, and the only thing that matters for the impact of the melody is the series of successive intervals (i.e. your Method 2 or what I refer to as the “distance” approach).

So which is right? Who knows? As with all things, the truth is probably somewhere in the middle. Alain Benbassat, who wrote the Functional Ear Trainer program Sarel mentioned, says that when he transcribes a song, he will usually start with the distance approach and then check the result with the functional approach. That sounds pretty reasonable to me.

Sarel, I can’t agree more that ear training is the route to better musicianship. I’ve been playing guitar for over ten years, but thanks to tablature, technology and laziness, I never bothered to train my ears. I resolved this year to change that, and my overall musicianship has improved immeasurably in a few short months.

Ron Gorow advocates a fonctional approach in his book wich you describe as method 1, and he discourages method 2.
The distance he encourages to be aware of is the distance to the tonic, wich is vertical distance and the same as the contextual approch.

drumdead10000
05-12-2010, 05:15 PM
check out this site:
its a flash application that gives youa random intervolt and you have to click on what one it is, its like a game, also i has the same thing but with chords also you can save your stats on a text file or something and paste them back and continue where you left off just do it like 10 munites a day and it will come to you eventually!!!!!

http://www.trainear.com/