Vocalist.org archive


From:  Dre de Man <dredeman@y...>
Dre de Man <dredeman@y...>
Date:  Tue Oct 17, 2000  1:33 pm
Subject:  Re: Technology (freq. graphs; long and boring)

Dear co-vocalisters,

1st: sorry for misquoting the title, but I don't have
the other emails here. Then: I used Gram, maybe did
something wrong, but did not see indications of
frequencies, so I went back to CoolEdit.

First I compared a recording of the tenor Hans Peter
Blochwitz and of my own voice, both singing the last
E4 (begin of passagio) from Schubert's 'Im Frühling',
and both with piano. (My recording was made last
Sunday). I chose this E, because it is a fermate note,
both Blochwitz and I make the same kind of (mf) tone
here and you hardly hear the piano in both cases.

The graphs look very simular. My recording was made in
very 'dry' (acoustically) room, so that explains
probably why there is more to be seen above 15 kHz in
Blochwitz' case. Blochwitz was also recorded closer,
not unlikely with a Neumann tube mike that will
produce some soft extra harmonics and he is singing a
bit softer, at least on the beginning of the note, if
I am allowed his breathing noise as a reference. (It
sounds quite loud on his recording).

Note that both Blochwitz and I don't have a peak at
4kHz, but at 2.5 and 5! When I first saw only my own
graph, I thought there was something wrong with my
tone production, but since Blochwitz's graph looks the
same, I think it might be ok. Besides that there is no
dip at 4k, just no peak.

I also tried to make a few other graphs, from notes I
just sang for the graphs. That was a disaster: I
forced my voice to make a high c that would make a
graph like the ones in Miller's book, but it did not
work: the fundamental was (slightly, but still) lower
than the first and second harmonics, and beside that:
it just did not sound good, metal was lacking (it was
loud though).

Then I made one graph of messa di voce on a G4 and
that was interesting: the pp part still had high peaks
around 3 and 4 k, that explains why such tones carry.

So what did I learn, after all?
Nothing. Although the graphs of Blochwitz and me look
very simular, you can clearly hear differences, so my
ears tell me more than the frequency analysis. And
although Blochwitz has sung in La Scala, I don't think
sending them my frequency graph will make them hire me
(the'd better not!)

I did not learn much from the other tests either. Even
that the ppp carries I should have known: my pianist
tells me so every time when I am in doubt whether to
do a morrendo or not. Besides that I've heard it from
Cecilia Baroli's concert in a huge hall two weeks ago:
she sings quite soft to begin with ( a little bit too
soft, I think ), and then makes diminuendo's and
sometimes morrendo's: you still hear it very well on
the last row. (Though it was so soft that the violist
was not able to make a tone that was less than about 5
times her volume; I even pittied him a little, because
his tone was falling apart a bit.)

Conclusion:
1. No loud high c's for me for now.
2. What helps me more than frequency graphs, is to
listen to my own recording, and of course what other
people tell me.

So to conclude this email that is much too long
anyway, I will tell how I make my recordings.

The recordings of my practice sessions with my pianist
are made in a small room with two microphones at the
same position, about 3 meter (9 feet) from me and the
piano, at an angle of 90 degrees from each other,
using a Sharp MD. (When practising alone in a simular
room, but then with a CD-recorder with good AD & DA
converters. (The DA converter I use also to play the
MD's back.) In this way you record the sound exactly
like it sounds, apart from the MD datareduction. The
microphones record every frequency between 20 and 20k
Hertz equally loud and also don't colour sounds from
the sides. Radio recording are sometimes made in this
way (but in an acoustically much better environment),
CD recordings never: the voice is recorded more close
then, which makes it sound warmer and bigger. The
piano can be recorded more beautiful and clear as
well, putting the microphones closer. But that is not
my purpose: I use the recording to hear as critically
as possible what I am doing. That in a concert I will
sound better, is nice of course, but now
I want to hear everything I do wrong as clear as
possible! Also the loudness relation voice/piano can
be monitored very well in this way. (Although with my
pianist it is always ideal:)

Last but not least: I use Bowers & Wilkins DM6
speakers to monitor. Those are the speakers EMI and
other companies used 25 (!) years ago to monitor their
recordings. (I bought them used and refurbished them.)
I also have an exellent set of Sennheiser headphones,
but with them I hear less vocal errors and you cannot
hear the phase cancellation

Using my speakers I can exactly hear the smallest
vowel colouration, the slightiest mispositioning of
jaw or tongue, every nasality etc. But I still needed
and need other people to make me aware of all the
terrible things I have been, and partly still am,
doing wrong. I think human feedback still works the
best, if you find a teacher that is honest and
qualified enough to give you that. (Which I have
lately, although somebody who is quite bussy, but that
is another story)

Best greetings,

Dre

__________________________________________________

Attachment
E_ImFr.bmp
Type: image/bmp
Size: 59k
Download


  Replies Name/Email Yahoo! ID Date Size
5691 Re: Technology (freq. graphs; long and boring) saint james   Wed  10/18/2000   2 KB
5706 Re: Technology (freq. graphs; long and boring) Dre de Man   Wed  10/18/2000   4 KB
5709 Re: Technology (freq. graphs; long and boring) Sheila Graham   Wed  10/18/2000   2 KB
5714 Re: Technology (freq. graphs; long and boring) Dre de Man   Wed  10/18/2000   3 KB

emusic.com