How do you transcribe?

[mcmurray]

My question is what is your chosen method of transcribing music: Functional Pitch Recognition, or Interval Recognition?

 

from wikipedia;

 

Functional pitch recognition

 

Functional pitch recognition involves identifying the function or role of a single pitch in the context of an established tonic. Once a tonic has been established, each subsequent pitch may be classified without direct reference to accompanying pitches. For example, once the tonic G has been established, listeners may recognize that the pitch D plays the role of the dominant in the key of G. No reference to any other pitch is required to establish this fact.

 

Interval recognition

 

Interval recognition is also a useful skill for musicians: in order to determine the notes in a melody, a musician must have some ability to recognize intervals. Some music teachers teach their students relative pitch by having them associate each possible interval with the first two notes of a popular song.

 

Just to be clear, when I say interval recognition I mean keeping track of every interval between every successive note, not relative to the tonic.

[GreenAsJade]

I voted functional pitch recognition, but really this is a lie.

 

I use absolute pitch recognition. I listen to the note and work out what note it is. I am not conscious of utilising any knowledge of the role of the note "in the context of an established tonic". I just listen to the note, ya know?

 

GaJ

[JonR]

I use both, and don't really make a distinction.

 

Or rather, I use the former (relation of notes to key) to identify chord sequences.

I use both but mainly the latter (relation of notes to chords and surrounding melody notes) to identify indiviidual notes of tunes or riffs.

 

I guess the simpler the music, the more I would use the former, because each note is likely to have a clearer relationship to the key than to the immediate context.

[mcmurray]

I thought this post sums up the differences perfectly;

 

http://www.ibreathemusic.com/forums/showpost.php?p=141992&postcount=2

[jonPhillips]

I listen to the note and find it on the guitar. I'm getting to the stage where I've transcribed enough stuff to begin recognising patterns I've come across before, but I suppose I'm probably recognising intervals. If that helps at all

[mcmurray]

 

I use both, and don't really make a distinction.

 

 

Just to be clear, when I say interval recognition I mean keeping track of every interval between every successive note, not relative to the tonic.

[JonR]

I thought this post sums up the differences perfectly; http://www.ibreathemusic.com/forums/showpost.php?p=141992&postcount=2

I think I understand the difference, but to me it's two sides of the same coin, if you like.

Any single note in a piece of music has various relationships with the others, by which we can help determine what it is (setting aside perfect pitch for the moment ):

 

1. To the key. This of course assumes there IS an overall key, and that the music isn't atonal (obviously!) or going through some kind of modulation at that point - although of course it wll relate to any new key that can be perceived.

 

2. To the root of the current chord. Again, we're assuming there is a chord, with an identifiable root. Obviously in unaccompanied melody, there is only the key to relate it to (as its tonal environment). But as long as there are harmonies of some kind, the note has vertical relationships with the other ones simultaneously present.

 

3. To melody notes before and after. As we listen, of course, we only have "before" to guide us, but we can shuffle back and forth when transcribing phrases in a recording.

 

1 and 2 are - as I see it - harmonic (vertical) relationships, and 3 is a melodic (horizontal) one. 3 can be above or below, while we always measure 1 and 2 from a lower note (key tonic or chord root).

 

I'm wondering if perhaps what you're calling "functional pitch recognition" is what I've seen elsewhere described as "harmonic intervals", while "interval recognition" is referring to what's otherwise called "melodic intervals" - notes in succession rather than simultaneous.

 

To me, that's an important distinction worth making. The way intervals work between notes of a melody - although we define them the same way - is a little different from how they work when in a chord, or as perceived in relation to the tonal centre of a key. The latter two relationships are a matter of "harmonic environment" - key (set of previously heard and remembered pitches) or chord (currently audible set of pitches). Thats still a subtle and interesting difference, of course: that of an actual harmonic environment (pitches currently sounding) and a remembered environment (the previously established tonality).

 

However you look at it, I don't see any one of these relationships as being superior when attempting to transcribe music. I would try and be aware of them all, to varying degrees. In a sense, I see them all as "functional"; certainly 1 and 2 are functional in the traditional sense.

 

(And I would work the same way when improvising, btw - although there I tend to favour 2 and 3 over 1. IOW, in transcribing, one is kind of reversing the process of composition or improvisation. In composition we work with all three things equally: find a melody, fix a key, harmonise with chords. In transcription, we pull apart (and rebuild) what someone else has put together, by listening for the relationships which they chose, in the same way in which they chose them.)

[JonR]

 

Just to be clear, when I say interval recognition I mean keeping track of every interval between every successive note, not relative to the tonic.

Ah, gotcha. (I posted the above reply before I saw this post) That's what I was guessing.

I still say both ways are of use, neither has priority for me.

[Virgman]

 

I listen to the note and find it on the guitar.

 

 

Me too.

 

I think this is called the "listenhearthenote" method.

[JonR]

I listen to the note and find it on the guitar.

Actually, this is in practice the way I work too. Of course, I hear various intervallic relationships (functional and melodic), but often I can't be sure enough of them - and sometimes one's assumptions about key (or familiar phrases you think you recognise) can mislead. There's no substitute, in the end, for zeroing in on individual notes - ignoring context as much as possible - and matching them on guitar (or keyboard or whatever). It's often the only way to be sure.

 

That's not a great way of ear training, of course! But generally, I'm transcribing in order to learn a song, not to train my ear. I reckon that happens all the time anyway, as I listen and play.

[Virgman]

Well, Jon, I agree that anytime we use our ears we are "ear training".

 

The ultimate goal would be to "play what you hear".

 

For me anyway.

 

Play as we talk.

 

I guess that's the blues really. Call and Response.

[fysicalgraphiti]

Let me try to get a hold on this, by Functional Pitch Recognition I think you're referring to identifying pitches in a tonal sense (a.k.a. Tonality). Looking upon a solo, melody, or chord in relation to the overall key.

 

While Interval Recognition is going about identifying intervals in relation to each specific chord change (a.k.a. Modality). Analyzing each chord or scale's structure relative to it's own root.

 

Is this that right or am I just going off on a different tangent.

 

Anyway when transcribing I really don't have a problem figuring out the harmony, melody, or solo. What the key issue for me is the rhythmic dictation when transcribing a blowing section. Right now I'm working on "Stablemates" by Miles Davis and his blowing section (originally by Benny Golson).

[JonR]

 

Anyway when transcribing I really don't have a problem figuring out the harmony, melody, or solo. What the key issue for me is the rhythmic dictation when transcribing a blowing section. Right now I'm working on "Stablemates" by Miles Davis and his blowing section (originally by Benny Golson).

Yes - rhythm and timing is a whole 'nother issue!

 

I find Transcribe is invaluable here. Usually I'm pretty good on rhythmic recognition (better than harmony anyhow), but to be absolutely sure, Transcribe's beat marker function is great: you can isolate single beats and check exactly where each note falls within it.

Of course, it's important not to get too anal about it, with jazz anyhow. You don't want to translate some lazy laying-back as some kind of precise 32nd note syncopation, or whatever. Allowance needs to be made for feel (swing), IOW.

In funk, it would be different. Exactly where those 16ths fall in relation to the beat is usually critical.

[mcmurray]

 

1. To the key. This of course assumes there IS an overall key, and that the music isn't atonal (obviously!) or going through some kind of modulation at that point - although of course it wll relate to any new key that can be perceived. 2. To the root of the current chord. Again, we're assuming there is a chord, with an identifiable root. Obviously in unaccompanied melody, there is only the key to relate it to (as its tonal environment). But as long as there are harmonies of some kind, the note has vertical relationships with the other ones simultaneously present. 3. To melody notes before and after. As we listen, of course, we only have "before" to guide us, but we can shuffle back and forth when transcribing phrases in a recording.

 

 

For the case of music that has a tonal centre, I think that if someone has a firm grasp on 1, then 2 becomes redundant. Harmony can be transcribed/improvised by listening for chord tones relative to the key, not to the root of the chord. This can be done quite easily with practice.

 

 

As you said though, both ways are definitely of use. With complex harmonies it's always good to have more than one trick up your sleeve.

[3shiftgtr]

For me, it is what ever works. Functional get-the-dang-thing-right-and-get-it-on-paper recognition.

 

Which often includes cussing and screaming at the wall.....which we will, for pedagogical purposes, call Functional "I SUCK!" Frustration Recognition.

 

Seriously, I just do what ever it takes. Even though I use both, it kinda goes beyond functional pitch and Interval Recognition. When figuring out the overall harmonic structures of a tune, I often keep an eye on the tonic...but that can be binding for some musics (bands like Periphery or Dillinger Escape Plan)...but an advantage for others (country, pop, etc). Generally tho' with chords in more complicated structures, it is a matter of just finding a handfull of intervals that work and going from there. YOu don't do either. If it is a guitar thing, I'll find a few intervals that work, and hunt the fretboard until I find a place where the whole chord will appear under my fingers. With 10 finger piano chords or horn stuff, I do the same, but look at the thing when I'm done and see how the stuff relates on paper, and that has fed me a few solutions. Stuff like "That sure didn't sound like a ii-V but it would make sense." Then I'll really try to listen inside the chord for the order of the intervals and try to figure out what the player/arranger was thinking, and that often leads me to something. So I use both, but for more complicated stuff, it is kind of both Functional Pitch AND Interval Recognition AND....um...neither....I guess.....

 

Melodically I keep an eye on both concepts as well. But sometimes it is the feel of the whole line...like you can just hear that it is a diminshed line over a dominant chord....so which concept is that? Both? Example: I'll hear the 'color' of the line....like I'll hear a guy play a line I recognize as Mixolydian, then I'll hit the fretboard thinking mixo in the key or over the chord or whatever and poke around till I find it....different fingerings and such....I'll listen for passing tones or other tonalities. Now in the jazz thing, these line 'colors' get more dense, and often alter the flavor of the chord under it and vice versa. Then it becomes a mental game with the player. "Gee, that line sounds minor-y, I wonder if he is doing the Pat Martino thing." Or "Man that line sounds Mixo, but it doesn't fit with the harmony...but it sounds good....secondary dominants maybe?" Recognizing these concepts and using that as a spring board to find the notes; which is that Functional Pitch or Interval recognition?

 

And sometimes I just go note to note and don't think about it's relation to anything until i'm done, though I might recognize a few things along the way. Then I'll do an analysis to see what they are thinking. That's how I grew to love Wes Montgomery.

[Jasco]

I'm from the 'whatever works' camp too.

 

I never really categorized or analyzed how I transcribe.

 

Hunt and peck for notes on the fretboard, sing them, listen for chord qualities, intervals, key-center relationships, similarities to previously known songs, applied theory knowledge, trial and error, lucky guesses, ect...

[BydoEmpire]

 

I listen to the note and work out what note it is. I am not conscious of utilising any knowledge of the role of the note "in the context of an established tonic". I just listen to the note, ya know? GaJ

That sums it up for me, too. I voted Functional Pitch Recognition, but it depends on the part.

 

For certain things I relate everything to the root (e.g this sounds like a b3 V b7), but for others - particularly faster melodies - I'll just follow the intervals and go by ear that way (sounds like 1/2 step, 1/2 step, whole step, major 3rd up, etc).

[jonfinn]

When I transcribe a solo, I usually approach it this way:

 

1) Get the bass motion first. It usually gives strong clues to harmony (key, tonal center etc.)

 

2) Get the chords. Chords don't automatically follow bass motion. It's good to check. Use that information to figure what scales, arpeggios, melodic pitches might be likely. But never assume anything.

 

3) Do an analysis of the chord motion and check if there are any "out of key chords". Make an educated guess as to the reason for the out-of-key chords. Note that for melodic possibilities.

 

4) Transcribe a "rhythmic outline" of a section of the piece (usually starting with the hardest part). Notate rhythms but not the pitches quite yet.

 

5) Note any "stress points" within the melodic movement. Find the pitches of those "stress points"

 

6) If I "just know" what the rest of it is, then I notate it. If not, I start making educated guesses.

 

7) practice what I've written down exactly as I've written it (not how it sounds on the recording). Then check that against the recording. If I'm wrong, make adjustments.

 

8) Keep making corrections until I'm unable to detect any further discrepancies.

 

For me it's a "try to understand the big picture" thing.

[BydoEmpire]

When I transcribe a solo, I usually approach it this way: 2) Get the chords. Chords don't automatically follow bass motion. It's good to check. Use that information to figure what scales, arpeggios, melodic pitches might be likely. But never assume anything. For me it's a "try to understand the big picture" thing.

That's really good - I've been slowly getting myself to do that. Out of bad habits or ignorance or whatever I tend focus on trying to get the melody or bass line first without understanding the chordal structure. Maybe because the melody is what you "hear" first, but it's invariably a lot harder to do that way. I fall into a similar trap improvising as well. It's fine if the chord progression is simple, but trying it "wing it" over more complex changes simply doesn't work. What's that golden rule of communication - seek first to understand?

[jeremy_green]

Getting the chords first is always my first step. I find that helps IMMENSELY!

 

I just don't trust bass players : )