Players' Vibrato Database

Pete Thomas

Well-Known Member
Cafe Moderator
Commercial Supporter
I have started a knowledgebase of different players' rate of vibrato. Is this a sign of old age or just having too much time?

Actually I'm using it as part of TamingTheSaxophone volume 1: Tone Study.

I was quite surprised to find Johnny Hodges and King Curtis exactly the same. Those are the only two I have so far, but if anyone cares to send in any info it will be gratefully received and appreciated. It needn't be too accurate, but if you have a way to measure vibratos per second or per beat at a certain tempo, please let me know.

Looking for any well known player.

Some are quite tricky, e.g. Coltrane has mostly no vibrato, just a wobble or two on the end of a note sometimes.
 
Pete - judging from the avalanche of replies you've received it would seem that you've stumped us all.

I just listened to some bits of Dexter Gordon's playing to see if I could "take his pulse" and my initial conclusions are that it would seem to be a maddening task. I put "Cry me a River" from "Dexter Blows Hot & Cool" through Transcribe! at 25% speed and he seems to play about 3 - 3.5 vibrations / second when he holds a long plaintive note, but I couldn't swear by that number. It's also rather a difficult thing to listen to!

Bill
 
I agree, I tried with Dexter and Coltrane, there's just a wobble at the end of some notes so impossible.

Plus, to really count you need some recording software, I wasn't expecting a lot of people, I posted just in case.

I'll be able to do Bechet though, and hopefully Kenny G.
 
Turns out that this is bizarrely interesting. Just listened to Coleman Hawkins playing "Laura" from "The High & Mighty Hawk" and he clocks in at nearer 5 Hz. I always figured he'd be a lower frequency player than Dexter!

Pete, the question that comes to mind is how elastic, or otherwise, is a player's vibrato? If we can generalise that is? In other words, most of us use at least some vibrato in our playing, but does the frequency of vibrato tend to be the same regardless of the tempo of the piece, or do players tend to use a faster vibrato with faster tempos?
 
I agree, I tried with Dexter and Coltrane, there's just a wobble at the end of some notes so impossible.

My teacher once spent some time trying to teach me to imitate lester Young's vibrato (on a fast tempo transcription).
It is a strange effect where it becomes narrower at the end of the note, kind of fading out. I couldn't do it, anyway.

It s likely to be a hard challenge... are narrow, wide, fast, slow the basic parameters, or you have some other ideas?
 
This will be interesting and useful.

I wonder what technology there is that could help us extract information from recordings. Reading from your thread on SOTW it looks like you have something that can draw a good frequency vs time trace from what is presumably a monophonic recording of your sax sound. But when it is a band recording with lots of other instruments in the way, it would be more difficult. I have an iPhone app called Pitch Primer that could work with single instruments - maybe that is what you have used ?

What measures would you use to characterise vibrato ? Maybe rate of vibrato, range/depth and pitch centre ? But there is also when it starts - e.g. constant or just added towards the end of long notes.

I guess that individual players may have different vibrato for different tempos, musical genres and at different times in their careers.

Other sax players I would be interested in would include Harry Carney, Cannonball Adderley, Johnny Hodges, Ben Webster. I guess that Sonny Rollins has an enormous range of different vibratos, from none to an exaggerated one - I must look for examples.

Rhys
 
What measures would you use to characterise vibrato ? Maybe rate of vibrato, range/depth and pitch centre ? But there is also when it starts - e.g. constant or just added towards the end of long notes.


Rhys

Rhys - wavelength and amplitude. I think Pete should actually graph the various sine waves of these players' vibratos.
 

Basically an FFT of the squared signal.

Slightly less basically:

Band-pass filter the input signal to the frequency of the sound that's being modulated (Last time I was looking at broadband noise, but in this case, I'd filter it down to the pitch of the note in question).

Square the signal. This is the half-clever bit that results in the FFT giving you the amplitude of the modulation, rather than just the mean amplitude of the tone.

Run the data through a low-pass anti-alias filter.

Down-sample so your bandwidth goes from a few kilohertz down to, say, 20Hz.

Then an FFT will give you the modulation spectrum.​

Like I say, I don't know how to go about dealing with frequency modulation, but I'd welcome any suggestions!
 
When we humans try and quantify art things get very complicated, and very amusing.

Just look at all the studies on violin sound characterisitics. Chladni patterns, sonic analyses.....

Don't get me wrong ..... I wish you all the best in this ..... Its just one thinks one is getting close and then some bright spark says BUT what about .......
 
Last edited by a moderator:
What about the difference between jaw and Diaphram vibrato? some players use both.

That is almost the point of pitch/volume vibrato.
With jaw vibrato you tend to go flat/sharp on the note (a software tuner could detect its amplitude)
With diaphragm you are supposed to vibrate the volume of the note, without altering the pitch. Not easy.

I tried to check my own vibrato: not an easy task.
Apparently I start vibrating long notes after one beat, to let the note die. I think I picked this from some early Coltrane, but not sure. I vibrate on triplets of crochets or quavers, if the tune is slow enough
 
Basically an FFT of the squared signal.

Slightly less basically:

Band-pass filter the input signal to the frequency of the sound that's being modulated (Last time I was looking at broadband noise, but in this case, I'd filter it down to the pitch of the note in question).

Square the signal. This is the half-clever bit that results in the FFT giving you the amplitude of the modulation, rather than just the mean amplitude of the tone.

Run the data through a low-pass anti-alias filter.

Down-sample so your bandwidth goes from a few kilohertz down to, say, 20Hz.

Then an FFT will give you the modulation spectrum.​

Like I say, I don't know how to go about dealing with frequency modulation, but I'd welcome any suggestions!

Way beyound my level of understanding. :w00t:
 
I wonder if a recording occilliscope would do the frequency analysis piece. I will have to look at my old scope to see if it has the necessary output for recording.

Freq on Y time on X. Be a visual thing but if itncan be done on a scope well then some bright computer geek can create numbers.

Have to notch filter really tight tomthe fundamental but hey mYbe it could work.
 
Last edited by a moderator:
Having been an electronics engineer for many years I actually understand the talk of square was established and frequency modulation also having held an amateur radio operators licence for many years there was a time many years ago that I may of been capable of analysing a frequency, however I fear that may not be the case now!

However being a mere novice at playing saxophone I do not know what vibrato is? Apart from the make of a rather arty looking plastic sax that Sax.co.UK sell!

Simple logic would steer me towards it being a word to describe vibration of some level!

Pete if I could ask you or anyone else for a definition that a beginner has a chance of understanding is be grateful.

Rgds Lee.
 

Members' Blogs

Trending content

Forum statistics

Topics
29,587
Messages
513,059
Members
8,736
Latest member
Fitzon_cider
Back
Top Bottom