chord voicing ear training

[abminor]

Hi,

Here's my problem: I would like to be able to figure out the correct voicings when I transcribe a piece of music but most of the time I have troubles to do so. Generally, I end up with over-simplified version of the tune where most of the chords are functionnally correct but do not have the correct voicings (sometimes it's even worst because I may mix up chords that are harmonically equivalent). I find that while the harmonic progression is preserved I loose most of the beauty of the tune because of that.

To make if short: How do I find out the real notes payed when transcribing instead of relying on my approximate harmony sense.

Do you have any tips to become better at it ?

Thanks.

[abminor]

For instance,

I'm currently trying to transcribe this arrangement of "when sunny gets blue": http://home.mindspring.com/~b.kane/s...nyGetsBlue.mp3

I don't know if I should proceed by listening to individual voices or rather listening to each chord voicing as a whole.

With the first method, I can already find the melody and bass lines but then I have troubles to hear the inner voices.

With the second method I'm more or less able to find the chords (althought somes still cause me problems) but I'm really not sure the voicings are correct.

It would really be helpfull if some of you could describe to me the process they use to transcribe this type of tune and particulary how do you find out the inner voices.

I was thinking also, maybe this type of transcription is above my level at this moment. Would you have some easier tunes examples I could start with, to practice transcribing music with several voices.

I read often that starting with bach two part chorales, then three part, and finally four part is good to practice harmonic aural skills. Do you think doing that would make me able to transcribe the kind of piano coktail tunes I wish to play.

[CrazyTalk]

In my opinion, being able to identify chord voicings is a case of spliting the different tones apart and hearing them for what they are. For example, if a guitarist was to play the chord 'x7678x', at first I would hear the chord as a whole, then break it down into its lesser parts. A root, a Major 3rd, a Tritone and a Perfect 4th, you alredy know the root is E, the rest is just knowing your intervals. a 3rd from E is G#, a Tritone from G# is D and finaly a 4th from D is G. So the chord is played EG#DG, doing that at speed takes practice. I'm sure there are other ways of doing it, perhaps even more efficient ways, but thats the way I understand it. By the way, thats a really nice piece of music!

[bluesking]

This is a problem I usually suffer as well. I have found a system which works for me and perhaps it could help you too.

Getting the overall harmony as you already have is a good starting point. The next thing to do is to find the lowest voice (usually what the bass is playing) and the highest voice (lead instrument, lead fills, even voice sometimes). Then find voicings for each of the chords you identified with the correct lowest & highest voice. This gets you very close to the overall harmony of the tune and keeps the voice leading relatively correct. Any errors which this method creates (i.e. in the middle/inner voices) are not particularly important because the practical options for guitar chord shapes cause just as many issues (its impossible to play many piano inversions without great difficulty for example).

Another point of view which helps is to concentrate on guide tones. This can often help co-ordinate the inner voices.

In my experience, each voice is likely to contain the following types of note relative to the chord of the moment:

Bass voice: "Perfect" intervals (Root, 2nd, 4th, 5th)
Inner voices: Guide tones (3rd, 7th)
Upper voice: Extensions (9th and above)

If you are playing solo your responsibility is increased because you need to handle as many voices as required to keep the feel of the overal tune. If you are comping with a bass & soloist you should usually just use guide tones to keep out of the other musicians' way. If you are soloing you can get a lot of mileage out of extensions because, once again, they keep out of other musicians' way.

These are not firm rules of course, but they can help you arrange voices if you are having problems. Also, these rules probably only apply in a Jazzy/Bluesy context. They wouldn't necessarily work in metal (I don't really understand metal 'harmony' too well, if there is much to understand).

[abminor]

Thanks for the tips both of you. I'm glad someone finally answered my call.
I have to say I a temporary delayed the transcriptions I was working on and moved to easier ones (easier voicings).

CrazyTalk: I will give it a try but I have always be bad at hearing verticals. Appart from basic qualities like major minor and diminished, I have a really hard time identifying chords and even more correct voicings that way.

I have a more horizontal hear I guess. I tend to recognize chord because I recognize the progression, not individual chords.

bluesking: Thanks for the practical tips for a narrowing the possible voicings choices. That may help.

[CrazyTalk]

By the way, there are two decent free ear training sites that I know of. http://www.musictheory.net/, and http://www.trainear.com/. There are also two cool ear training programs, which are free. Functional Ear Trainer at
http://www.miles.be/, and Solfege http://www.solfege.org/. All these applications are great for ear training. Another really good package that is NOT free is http://www.earmaster.com/.

[JazzMick]

Assuming piano is your instrument.

Simply put it requires time and experience to fully hear the voicing's while you are transcribing them.

You could sit down with a tape recorder for hours every day listening to millions of voicing's and attempt to recognize them by ear. My recommendation would be to just learn tunes.

Always strive to expand your knowledge of voicing's and voice leading. Substituting chords and so forth. It takes years but if you persist and be realistic about the music you want to transcribe you WILL improve. More than you might expect.

As a thought. Maybe try and balance part of your practice time between 1) transcribing something within your ears range. 2) transcribing something a bit tricky. (it can be as little as 4 bars) then put it away and come back a week or two later to see if you can find mistakes or improve on it.

3) Get some accurate transcriptions of tricky tunes that use nice chordal movements. Either buy some books or find them online. Doing this is worthwhile. It saves you the trouble of having to figure it out but still gives you the insight into what those sounds actually are. What the chord names are and what notes are being used for each particular voicing.

Anyway, Good luck! Remember, If you haven't played it before... you will probably find it hard to transcribe. So explore voice leading as often as you can when practising.

PS. Don't be put off if you cant always transcribe something 100% accurately. Often I like to just get the leading tone movements and then fiddle around with which ever voicing sounds best to me at the time.

[fingerpikingood]

what i do when i want to ear something out with correct voicings is stick the tune inside of software that allows me to slow it down and loop it. usually just looping it works well enough. you can pick out the notes one by one that way.

if you can get your hands on something like adobe audition formely cool edit pro, or soundforge or something like that this is really good for that.

also KMplayer, my media player of choice which is free, i believe allows you to loop sections.

it's kind of like winamp except it plays everything, from internet video feeds (standard ones are all korea tv cause it's korean) DVDs to mp3s. it'll play all video formats. no need for quicktime and real, and this and that. and like i said i think it will do the slow down and looping. it really has a hell of a lot of options with it. but it looks like winamp kind of style GUI.

whereas adobe audition typed software you see the waveform and you can highlight sections of it and loop those. easier way.

software like transcribe might be useful too you as well. it attempts to ear stuff out for you. but it's not perfect because it doesn't always know which transients are the harmonics and which are the actually resulting notes you hear. i think this may be free as well. or if not it must be cheaper than something like adobe audition.