For Gerald Eskelin: Timbre of Singing (was Re: Jerries: a conclusion > or two)

On 3/18/02 3:16 AM, "tuning@yahoogroups.com" <tuning@yahoogroups.com> wrote:

> Message: 17
> Date: Fri, 15 Mar 2002 23:58:50 -0000
> From: "jacky_ligon" <jacky_ligon@yahoo.com>
> Subject: For Gerald Eskelin: Timbre of Singing (was Re: Jerries: a conclusion
> or two)

> --- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:
>> Simply stated (I hope), we're looking for the "exact" relative frequency
>> that will sound "in tune" with what we (I, at least) believe is an
>> illusionary "high third" produced by finely tuning the major triad in JI.
>
> Jerry,
>
> I hope you won't mind if I offer a little input on this topic.

Are you kidding? I'm delighted. We can use all the help we can get.
>
> It is my opinion that timbre plays greatly into this perception, and
> *one will be deceived by doing these experiments with anything other
> than the human voice, such as synthesized tones.*

You're not the first on the list to have suggested this, Jacky. And, if we
hit a wall with the synthesized jerries, that's the first place we'll go.
Live strings is another possibility.

Last year, I posted some vocal examples of the high third. As I remember, we
didn't have a any offers to put these on a spectrum analysis (or whatever).
That may also be a way to go.

Actually, I think we're making some progress with the synthesized jerries.
At least we are getting some agreement on which are the "best."
>
> The human voice is continuously variable - changing moment to moment.
> One can see this if they record the wave form and "look" at it in an
> audio editor, as compared with a synthesized timbre. It's two
> entirely different sound worlds.

I've already noticed that when I dropped Paul's examples into my LogicAudio
software. Very strange looking waveforms. Symmetrical and sterile looking.
>
> I believe if you really want to find out what this "exact" relative
> frequency" is, you'll will have infinitely better perception of this
> interval/effect by using samples of real human voice. One could play
> these intervals on a digital sampler. I would suggest making some
> recordings of long held tones (this is how I frequently work), then
> experiment with different intervals until you find out what it is.
> With long held tones, rather than looped ones, one will better
> emulate the sound of actual singing. I don't think you'll find it
> (easily) referencing with a synthesized waveform.

I'm not sure "easily" is a term that's going to be any more applicable to a
"nervous system" pitch than to a synthesized one. But then, who knows what
may turn up.
>
> If I weren't so busy, I'd help you out, but I'm slammed on so many
> levels right about now, but I do find this interesting.

Okay, Jacky, you're off the hook. You just have to agree to buy the book
when it comes out. :-)

(Privately, I'm hoping you can't sleep at night thinking about our search
for the high third and feeling guilty about not getting involved. ZAP! ZAP!
ZAP!)
>
>> Regarding mp3, I'm not the guy to comment. My understanding is that mp3 is
>> sorta "stripped down" for speed. It may be subject to "categorical
>> perception." (Do you like that, Paul?)
>
> MP3 in my opinion will distort such an experiment as much as trying
> to base the results on listening to synthesized timbres, if one does
> not use a very high birate. 192 kbs, perhaps as a minimum.
>
> Good Luck,
>
Good luck accepted. Thanks.

Jerry

--- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:

> Last year, I posted some vocal examples of the high third. As I
remember, we
> didn't have a any offers to put these on a spectrum analysis (or
whatever).

you have now!!! from our friend from france (?) . . .