View Full Version : Ear Training and Musical Memory
bcrsf27
06-03-2008, 12:11 AM
Howdy Folks,
So here's a question for all the ear training gurus out there. How does one improve musical memory? I've been working fairly dilligently on ear training for the last 6 months or so and made some real progress. I've used EarMaster to drill intervals and Functional Ear Trainer (i think its called) to drill harmonic recognition. And I do pretty well in the context of these software packages (usually near perfect). But, when I actually try to identify a line of music of any length, I find that I can't hold the earlier pitches in my head. It's like I've got space in my music memory for 2 or 3 pitches and as new ones enter the older ones are pushed out. It's kind of frustrating because, for all the work I've done, it isn't practically useful. More of a parlor trick which, while cool, isn't necessarily what I'm after.
Anyway, I wonder if anyone else has had this problem or might have an idea on how to break through this current plateau.
- bcr
Don't really understand the problem. What occasions would require you to hold more than 2 or 3 earlier pitches in your head?
The purpose of interval recognition, IMO, is to be able to identify chord types and chord changes (for transcription or improvisation), and to assist in melodic composition. Once you identify them - one by one - you play them, or write them down. No need to hold many previous ones in your head.
I mean, I can't do that, but it's never been a problem.
(Perhaps I'm misunderstanding...)
And I don't think the human brain can do more any way. I've read that the maximum number of notes the brain can hold on to is 7, plus or minus 2 notes depending on individuals. That's from studies on attention span. Two or three pitches in the context of actual music is much more complex than 7 +/- 2 notes, so I wouldn't worry.
bcrsf27
06-03-2008, 09:38 AM
I think I need to do a better job explaining what is going on for me. Let's say I'm trying to transcribe a melody line is 12 notes long. If I listen to the whole phrase, I'll probably be able to figure out the last 2 notes without a lot of difficulty (or at least ID the interval between them). But to pickup the previous 2 I would have to stop the tape before the last 2 notes get played (or set the loop or whatever for the technology in question). If I don't, I loose the pitches. So maybe a better way to describe my problem is that my ear is easily distracted. If too much auditory information enters my brain after the information I'm trying to identify/transcribe, I lose the pitch information.
What's interesting is that this isn't true for rhythmic information. I can recall long rhythmic phrases quite easily (I understand this is a more widely distributed skill). It's just the pitches get lost.
The result of this is that I end up using my Transcribe tool to isolate 2 notes at a time (usually containing the note I have already IDed and the new one) which works but is incredibly tedious. Also, in live settings, it's tough to pickup a phrase that someone else plays (since they aren't necessarily going to have the luxury of breaking the phrase down for me while performing). I know there are people who can hear a phrase of music and simply play it back verbatim. Isn't that ideally the goal of ear training - to be more fluidly connecting our ear and our instrument?
Ooogie
06-03-2008, 11:40 AM
What you're describing is the nature of transcribing (listen, write, rewind), at least that's the only way I've ever been able to do it. I usually start writing the first couple of notes down and keep going until I lose it then restart the loop in ASD. The good thing about tools like ASD is that it's easy to move the start and end points of the loop.
I do try to use shortcuts whenever possible though. For instance, if I'm listening to something I'm real familiar with it's easier to pull out licks and move a little faster without having to hear every note. I also try to use my knowledge of the theory behind the playing to figure out the notes without necessarily hearing every interval. When I'm finished I always go back and play through it with the recording to correct any mistakes anyway and they tend to stand out more at this point since it's so ingrained in my mind.
Mark
bcrsf27
06-03-2008, 02:52 PM
... I usually start writing the first couple of notes down and keep going until I lose it then restart the loop in ASD....
Mark
Me, too. The problem (such as I see it - maybe it isn't really a problem) is that it requires that I literally isolate the 2 notes I'm working on - the note I know and the next one. If the loop includes notes before or after the notes I'm trying to identify, I get confused and feel I might as well be guessing. It sounds to like you take a music phrase, loop it, and transcribe a couple of notes from it on each pass. It may take many passes, but it is still picking out notes amidst a larger musical idea. That doesn't seem to work for me.
Fortunately, the technology exists to set very accurate loops such that I can still slowly get work done (not to mention the capacity to slow things down w/o pitch shifting). Just seems that there's something kind of weird going on for me. I mean, were it not for Transcribe (or programs like it), I wouldn't be able to do this at all. Not a single note. That's seems odd and somewhat difficult to reconcile w/ the fact that interval ID in isolation is pretty easy.
Any further thoughts based on the above? I appreciate everyone's insight so far, I think I'm just having trouble describing my experience.
- bcr
spencerbk
06-03-2008, 09:24 PM
First off, here's a great website that quizzes on melodies from 1-15 notes long
http://www.iwasdoingallright.com/tools/v2_23/ear_training.aspx
Second, if you can sing "happy birthday" then you have a musical memory of more than a few notes. If you'd never played it before and were working it out on guitar, you'd probably find your first note and then progress in familiar 2-3 note passages until you made it to the end. If your interval ID is really good you might be able to do that at a reasonable tempo. I don't think you need to visualize the whole melody naming all the intervals from start to finish before you pick up the instrument.
What I'm trying to do w/ Transcribe is learn to sing the bit I'm trying to learn, rather than find a few notes at a time and work them out on guitar. From there I'd be able to use my interval ID skills to find the notes based on my singing in the manner described above. Memorizing complex melodies is harder than learning new verses to Happy Birthday, but I bet practice helps ...
Ooogie
06-03-2008, 09:47 PM
Any further thoughts based on the above? I appreciate everyone's insight so far, I think I'm just having trouble describing my experience.
- bcr
I'm not sure what you're transcribing but try to use all the clues available to you when figuring out a song, don't work on the intervals in a vacuum. If you know the key, scales used and something about the artist's style it can really help narrow down the note possibilities. Also, if you're holding the guitar and playing it as you go some notes will just make sense while others would be almost impossible to play. Think like a detective ;)
Mark
Another thought is to do with "phrase memory".
I find I'm pretty good at remembering the sound of certain phrases, of at least 5 or 6 notes - esp in the blues idiom, which is what I've been playing for longest. IOW, I have a vocabulary, a library, of phrases I can consult when transcribing.
Trouble is, it can be a hindrance. I think I recognise a phrase, but when I zero in individual notes, I find I have it wrong. The phrase shape will be the same, and maybe all but one of the notes will be right. But my prejudice makes me hear the phrase I know and not the phrase that's there. I still need to zoom in on individual notes to be sure.
Judging from other responses here, it seems as if your "problem" is everyone's general condition. Par for the course. Yes, transcribing accurately is slow, and we have to often (if not always) have to go note by note to get it right. This is especially the case if we are experienced in a particular genre, where certain phrases crop up regularly, but not always played identically.
I use Transcribe now, but I used to use a 2-speed tape recorder - going back and forth over 1 or 2 notes ar a time to make sure I had it right. As soon as I tried guessing beyond 3 or 4 to save time, (it seems) I'd inevitably get it wrong.
Of course, experience (and theoretical knowledge) will also let us make educated guesses sometimes. Eg, if we know about key, we can soon spot notes that are outside the key, we can predict that a note will be one of a very small number of possibilities. But (for me) it still needs that note-by-note confirmation.
It can also save time to be aware of structure and repetition.
Here's a question: how long does it take you to transcribe an average song - say melody and chord sequence? (all sections: verse, chorus, bridge, etc)
For me it varies a lot, but I'd say a minimum of one hour. That would probably include the odd guitar riff or bass line. And it includes notating it (in Sibelius). But many songs will take a lot longer than that. It's not that the process is difficult, it's just time-consuming.
(Chord sequences alone will be a lot quicker of course.)
I dont find this a problem, or frustrating. I enjoy the process. It's like dismantling a beautiful machine or artefact piece by piece to see how it's put together. I'm slow and careful out of respect.
(Readers with perfect pitch will now be grinning smugly, I suspect... ;) )
bcrsf27
06-04-2008, 01:05 PM
I haven't abandoned this thread - but I can't give a detailed response right now. Work and stuff. I will respond as soon as things settle down (probably tomorrow).
-b
Hi,
Some of my thoughts and opinion.
1)ET methods focusing on measuring intervals does not really help people who are not born with Relative Pitch.
2)Pitch memory and music memory is very important and many tests have concluded that it is innate. Read Diana Deutsch's "Psychology of Music" and it will answer a lot of your questions.
3)In my small musical world here in NYC, almost everyone I have worked with can hear a tune and play it back note for note after 1 chorus. Melody and changes, that is. So transcribing 1 note at a time is not par in my world.
4)My advice is to investigate a different avenue of ET. Individuals can always improve upon whatever nature gave them.
GovernorSilver
06-04-2008, 01:41 PM
1)ET methods focusing on measuring intervals does not really help people who are not born with Relative Pitch.
I thought Perfect Pitch was supposed to be the ability that few are born with and Relative Pitch was supposed to be the common one?
If only a tiny minority of individuals had Relative Pitch, then there wouldn't be so many great jazz sax and trumpet players. Not being able to distinguish a minor 3rd, major 3rd, and perfect 4th apart from one another, for example, would be a disastrous impediment to any prospective improvising musician.
I thought Perfect Pitch was supposed to be the ability that few are born with and Relative Pitch was supposed to be the common one?
If only a tiny minority of individuals had Relative Pitch, then there wouldn't be so many great jazz sax and trumpet players. Not being able to distinguish a minor 3rd, major 3rd, and perfect 4th apart from one another, for example, would be a disastrous impediment to any prospective improvising musician.
In my experience, absolute pitch is yes or no. And you can not practice to get it no matter what others might say/sell. You can practice to memorize pitches and this works but it is not the same thing.
RP is a complex subject because almost everyone has it to some degree or other. And RP can be improved with a good regimen/method. When I say people "born with RP", I mean there are some that are born with such a high sensitivity to pitch, except for the first note in the musical sequence, it is indistinguishable from AP.
Studies show this type of RP to be more scarce than AP.
Yes, you are right in that if you can't hear, in improvising music that would be a severe limitation. For me, to hear it is to play it. However, music has many components and hearing is only one of them. Many so called "great" players are not really great. Some have technique and others think fast to cover for the fact that they can't hear instantly. Most end up playing music where hearing is not so crucial.
And I must disagree with you on "many great jazz sax and trumpet, etc." In my opinion, there is only about 1000 great jazz players on the planet at this time. Not even 1 % of all that try.
1)ET methods focusing on measuring intervals does not really help people who are not born with Relative Pitch.Everyone is born with relative pitch, at least at a crude level (easily built upon). Do you mean perfect pitch?
Anyone unlucky enough to have no sense of relative pitch would not be interested in music (except perhaps drum music).
2)Pitch memory and music memory is very important and many tests have concluded that it is innate. Read Diana Deutsch's "Psychology of Music" and it will answer a lot of your questions.How is "pitch memory" and "music memory" defined?
3)In my small musical world here in NYC, almost everyone I have worked with can hear a tune and play it back note for note after 1 chorus. Melody and changes, that is. So transcribing 1 note at a time is not par in my world.What is the nature of your "small musical world"? What training and experience do your friends/colleagues have?
I've been playing music for over 40 years, mostly amateur, occasionaly (more so recently) with professionals. I've never met anyone who can do what you describe, after 1 hearing - assuming you mean the chorus played right through once, not phrase by phrase. Changes perhaps. Not an entire melody (that they have not heard before). Well, I guess there may have been one or two, seeing as I've not seen them tested in that way.
I suspect, after a lot of struggle and plenty of trial and error, a few might manage it - if they hadn't forgotten what they heard by then.
I don't doubt you. I just think you're in an unusual world. (Unless I'm misunderstading you.)
4)My advice is to investigate a different avenue of ET. Individuals can always improve upon whatever nature gave them.Agreed. Or perhaps can rediscover what nature gave them?
rockinrob
06-04-2008, 05:18 PM
I think I need to do a better job explaining what is going on for me. Let's say I'm trying to transcribe a melody line is 12 notes long. If I listen to the whole phrase, I'll probably be able to figure out the last 2 notes without a lot of difficulty (or at least ID the interval between them). But to pickup the previous 2 I would have to stop the tape before the last 2 notes get played (or set the loop or whatever for the technology in question). If I don't, I loose the pitches. So maybe a better way to describe my problem is that my ear is easily distracted. If too much auditory information enters my brain after the information I'm trying to identify/transcribe, I lose the pitch information.
What's interesting is that this isn't true for rhythmic information. I can recall long rhythmic phrases quite easily (I understand this is a more widely distributed skill). It's just the pitches get lost.
What you're describing sounds perfectly normal to me. In which case the goal (or technique might be a better word) is to hear the general quality of the line, pick out a couple notes here and there, and then think of the rhythm and fill in the rest. A lot of it has to do with being able to hear chords- this chord over that chord. For instance take the notes A C D over a Cmin chord- that could be implying a Dmin7 chord. You want to be able to hear that tonality. Once you get the tonality down you can use whatever other tools you have to get the rest. Like Ooogie said, there's a bit of detective work.
If you're doing it on the spot it gets harder, but that's not always a bad thing. Think of it like the kid's game telephone, where each person whispers to another, and at the end what was said by the first person is usually very different than what the last person said. We all hear stuff differently, and that's why one guy can play something that he thinks is Coltrane sounding, and someone won't hear Coltrane in it at all. It's all about how we organize music and ideas in our heads, and we all do it differently.
That's why when I talk about transcribing I often say it's OK if you don't get it perfect sometimes, because you're still learning something.
3)In my small musical world here in NYC, almost everyone I have worked with can hear a tune and play it back note for note after 1 chorus. Melody and changes, that is. So transcribing 1 note at a time is not par in my world.
Really? My experience is much more like JonR's. And I'm including the top guys I know and have played with (many of which are now in NYC and I'm sure you know of). I mean, if you're saying they can hear a tune once through and then get through it- know most of the changes and fake the melody, then yes even most of the cats around here can do that. But as far as having the ability to play it correctly and note for note after one listen- no way.
But as far as having the ability to play it correctly and note for note after one listen- no way.
You'll just have to come here and check it out for yourself. I've seen it over and over again.
By tunes, I am refering to standard type tunes with "singing" melodies. Not Donna Lee type of tunes.
How is "pitch memory" and "music memory" defined?
Pitch memory is the retention of the pitch function within the music. The music memory is the gestalt of the piece including form, feel, time, harmonic movement, etc.
bcrsf27
06-07-2008, 01:06 AM
I finally get a chance to get back to this thread, and things appear to have gone dark. Alas. Well, here's an attempt to revive things.
This thread has actually been quite useful to me because it is made me think a little bit harder about what skill it is that I'm actually trying to develop. In terms of transcribing music, I'm already using most of the techniques that have been discussed here: thinking about key and harmony beyond just measuring intervals between notes. The "detective" part of the process is fully in play -- that's the part I'm good at. And the tools are such that I can break down and identify specifically what is being played for nearly anything that I'm interested in as things stand.
But, I really would like there to be more fluidity about my musical perception that can be accessed when I'm actually playing with other musicians. Now, whether that looks like what Gene is describing -- perfect reproduction on one hearing -- or something less quick and precise it is immaterial to me. Obviously, I like to get as good as possible, but getting better than where I am now is the goal.
The hurdle seems to be that pitch information seems to decay for me very quickly. To show you what I mean, but me describe an experiment I was conducting today. While listening to my car stereo, at various points I would stop the CD player after a melodic phrase and see whether or not I could reproduce the phrase either vocally or in my "internal ear". What I discovered is that pitch information was extremely difficult to reproduce, whereas pretty much everything else (rhythm, phrasing, dynamics, etc.) was a snap. This is true even when I was relatively familiar with the phrase (although, it was admittedly easier than when it was completely new).
So, before I can even approach doing something like what Gene describes, I have to figure out way to strengthen my "short-term-memory- pitch-retention" system. The ear training work that I've been doing has been helpful in other respects, but hasn't done much to improve this particular part of my brain.
I guess when I wrote this I was hoping that someone would respond by saying: "oh yeah. I remember when I had that problem. I did _X_ _Y_ times per day for _Z_ months and now it's all better." Perhaps that was hoping for too much. I think I'll definitely try the web site spencerbk suggested and maybe some more of the exercise I described above. But, any other advice (or thoughts or comments) now that I've refined my goals would be appreciated.
- bcr
But, I really would like there to be more fluidity about my musical perception that can be accessed when I'm actually playing with other musicians.
Hi Bcr,
ET is a very complex and organic skill. Some have it at birth and others will never become good enough at it to successfully play improvisatory music.
That being said, unless you are completely tone deaf. By tone deaf, I mean you can't tell the difference between a A1 and B5, or you have received injury to your auditory cortex, you should be able to improve your relative hearing to what you have now.
Agian, if you are interested, I would read Diana Deutsch's books and studies. It can be a sobering read for people who are not so gifted with hearing music. Not everyone was meant to play pro basketball either no matter how much the individual wishes.
I was fortunate to be born into a family with great music everywhere as my mother started me on classical music at age 2. I tested and was accepted to a major music university where part of the test included hearing 16 bars of 4 part writing and having to write that down from 2 hearing. So everyone in this school can do pretty much what I talked of earlier.
My advice is do ET if you enjoy it and get pleasure from it. Otherwise, it will be very difficult as an adult to change the part of the brain necessary to hear music. Of course, not everyone needs to "hear" music as you can still have fun and play,etc.
Just my 2 cents.
vBulletin® v3.8.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.