I’ve noticed over the months I’ve been on this list mention of sound files people can download in order to hear the tone of a set of pipes perhaps for the purpose of deciding whether to purchase from a particular maker or something like that (stops to draw breath of air).
I was just wondering what value there is in sound files given that with a bit of playing around with the knobs and mouse, one can compress, enhance, and just about anything else to improve the tone.
I guess the only thing they are useful for is to hear if the things are in tune. But still…
I suggest one shouldn’t place too much faith in a sound file to decide whether or not a chanter’s tone is to one’s liking.
Just rambling…or should I say, procrastinating, before I go and mark 300+ exam papers.
I’m by no means a specialist in reed-instrument acoustics, but I would like to point out that the perception of tone quality lies also in the nature of the transient attack (that is, the first millisecond or so of the note), possibly more so than in the harmonic content of the note itself (which we could describe as the “timbre”). Of course, a nice timbre is important when considering an instrument purchase, and it is true that it can very easily be enhanced with some basic effects (such as chorus or equalization). However, the transient attacks and the transitions between the tones cannot be enhanced much, unless you are a signal processing guru. If an instrument is poor, no amount of filtering or processing will hide it. So when you are listening to the sound samples, pay attention to the staccato, the trilles and other note transitions.
This is interesting to me because I’ve spent quite a bit of time analyzing spectrograms of piping. I’ve differentiated largely on the basis of the strength of the overtones (from flute-like to buzzy).
In what way is the beginning of the note different? I’ve seen this but don’t believe I can hear it.
I know this is hard to pin down but any clues would be appreciated.
I think I hear what you are saying. I’m listening to Paddy Keenan’s CD 'Na Keen Affair" and while I know I can reproduce the sound of his chanter as heard on the CD by placing a thick blanket over my chanter when playing , the staccatos and pippity-pips certainly have a quality to them. Whether this is both a result of a well-made chanter plus good technique, I’d say both.
The comments “signal envelopes of a transient nature appear as vertical broad band bars in the spectrogram” and "musical instrument sounds (plucked strings, striked bars etc.) have a very short broad band attack " are very pertinent. Since these transient elements are very short, you would need to sample at a very high rate and use a very short FFT window to be able to differentiate between two instruments with similar spectral characteristics. Unfortunately, I don’t think it is possible to pin-point exactly what is “pleasing to the ear” in those broad band sounds since, for all practical purposes, they look a lot like the sound of a spoon hitting a saucepan or your finest china crashing onto the kitchen floor.
A fun exercise is to take two such instruments (say, guitar and piano or banjo and harpsichord), to have them play the same tune in the same key, and remove all the transients in the time-domain. The result is a totally bland music (from a subjective point of view). The specificities of each instrument is also gone: one instrument sounds a lot like the other.
Naturally, it is the art of the player to pluck the string, hit the hammer, apply pressure to the diaphragm or coordinate the fingers in such a way as to make those transient elements sound pleasing to the ear.
Yes, you can’t quantify “pleasing” but you can appreciate the differences. The presence of relatively strong upper partials, for instance, make for a striking, but eventually irritating, tone. By successively filtering these out, I was able to find the areas where a muffled, flute-like tone turned into a “clean” reed instrument, and progressed into “colored” sounding, and finally into the buzzy. It’s much easier for me now to catagorize the tone of a particular chanter having gone through this exercise.
This also seems to be why a C# chanter has a noticably different character of timbre than Bb. I went on to try to figure out the effect of bore and finger-hole undercutting on the strength of harmonics, but sadly I’ve misplaced my notes. I’m bummed that I never thought to filter out the attack transients. Oh well.
As you suggest, the air pressure combined with speed of getting the finger off the hole may have a lot to do with how the note is perceived, which gets back to why Liam O’Flynn sounds like LOF on Seamus Ennis’s pipes, in addition to his technique.
“A fun exercise” – at least I’m not alone in this world.
