MOTM

How do vocoders determine whether a sound (usually
vocal) is voice or unvoiced? I have looked at some
scheamtics, but I am not 100% sure what is going on.

What it is determined whether a sound is voice or
unvoiced, how is the vocal handled differently?

When making this determination, is it instantaneous or
is there a "frame" in which it is performed? Are
there ever problems with the signal fluctuating
rapidly between voiced and unvoiced? If so, what kind
of artifacts result?

Thanks!
Shemp

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

One easy way to extract voiced/unvoiced information (as I recall) is
to use a high-pass filter with a cutoff at 12 kHz or so. Plosive and
fricative sounds have more energy in the higher regions of the
spectrum, so when an S or F or P or whatever is spoken, the HP filter
passes the high spectra which is rectified and used to trip a Schmidt
trigger (threshold detector). This is used to control an electronic
switch, which switches from the default carrier to a noise source.
As soon as the unvoiced sound is over, the switch reverts back.

johnm

The switching from the carrier to the NS is performed
on the HP channel only, correct? Just want to make
sure I am not missing anything.

Is the 12 kHz cut-off based on designs you have seen
or is based on linguist analysis? IOW, how arbitrary
is that point?

--Shemp

> One easy way to extract voiced/unvoiced information
> (as I recall) is
> to use a high-pass filter with a cutoff at 12 kHz or
> so. Plosive and
> fricative sounds have more energy in the higher
> regions of the
> spectrum, so when an S or F or P or whatever is
> spoken, the HP filter
> passes the high spectra which is rectified and used
> to trip a Schmidt
> trigger (threshold detector). This is used to
> control an electronic
> switch, which switches from the default carrier to a
> noise source.
> As soon as the unvoiced sound is over, the switch
> reverts back.

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

Tentochi wrote:

>The switching from the carrier to the NS is performed

on the HP channel only, correct? Just want to make
sure I am not missing anything.<

The HP filter simply provides detection for the switch. If you
wanted to vocode the word "shemp," for example, and were using synth
strings as an input, the vocoder would detect the "sh" and switch to
internal noise as the input, then would fall back to the synth string
input for the "emp" portion.

On many traditional analog synthesizers, the octave
was denoted by "equivalent" organ pipe lengths. I
think the extant of the range is something like 64",
32", 16", 8", 4", 2", 1". Typically the range is some
like 16" to 2".

What is the corresponding frequency range for each of
these pipe lengths? What notes are the pipes
themselves tuned to? Is this the first note for that
octave?

Thanks!
Shemp

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

All right all of you trainspotters out there. This
one is pretty difficult. It is in the March 2003
issue of Keyboard.

Did you find it?

It is on the last page where they have been featuring
various studios.

Audio Playground Synthesizer Museum and Recording
Studio (http://www.keyboardmuseum.com)

Studio C, Microgroove

http://www.keyboardmuseum.com/st/micro.html

http://www.keyboardmuseum.com/st/microlist.html

In the pic in the mag it looks like they have added an
additional flying rack to the right of the MOTM rack.
I am NOT sure if there is an additional MOTM rack
there or not.

It looks like MOTM is slipping in somewhere every
month into the major mags. Way to go Paul!

These guys are the only retailers of MOTM that I know
of. But there website is grossly out of date. For
instance, they still have a picture of the MOTM-100
for sale. I guessing that they really don't sell very
much MOTM gear. I really can't think of a reason not
to buy directly from Paul.

--Shemp

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

--- In motm@yahoogroups.com, Tentochi <tentochi2003@y...> wrote:

Allen Strange's unparalleled tome, "Electronic Music:
Systems, Techniques and Controls" is still
available!!! There are at least 100 copies in the
warehouse and they will print more as needed.

I am not going to go into the unbelievable details on
how I obtained this information, so here is how to get
the book:

Call McGraw Hill's Direct Marketing Division at:

1.800.262.4729

Press "2" to order

Title: "Electronic Music: Systems, Techniques and
Controls"

They have this incorrectly listed as "Pol Electronic
Music".

They also have this incorrectly listed as a 1st ed.
It is actually a reprint of the 2nd edition.

Author: Allen Strange

ISBN 0-07-523219-7 (this is for the reprint)

Cost: $37.56

Handling: $2.95

Tax: $2.33 (this is for all states)

Shipping: via UPS (other methods available upon
request)

Ship from ZIP code 43004

Shipping weight 1.4 lbs

This is in stock and usually ships the next business
day after you call.

There is a note in their computers that says this has
to be ordered from the Higher Education Division.
This is not correct; Direct Marketing can handle it.

You must have a credit card (debit card) to place the
order. When you add everything up, the total will be
just under $50.

Happy Twiddling!!!!

--Shemp

P.S. Dave, please update this information on your
website when you get a chance.

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

Sound travels at a speed of approximately 1130 feet per second.
Therefore the wavelength of a "sound wave" at a frequency of 1 Hz is
1130 feet.
The speed of sound does vary with some conditions such as elevation,
atmospheric pressure, humidity and temperature.
If I remember correctly an open organ pipe resonates at the
wavelength equal to the length of the pipe. I think a pipe closed
at one end resonates at a frequency of either twice the wavelength
or half the wavelength, I can't remember which off the top of my
head.
A 64' pipe would resonate at 17.65 Hz.
A 1' pipe would resonate at 1130 Hz.
You can make an organ pipe resonator by mounting tubes cut to the
proper lengths in front of a speaker. The notes corresponding to
the resonant pipes will ring producing a very nice reverb type
effect. (Something I built out fo PVC and copper pipe in the days
of my misspent youth.)
One other interesting bit of organ trivia.
Typically you would have multiple stops open at any given time.
Every note has a corresponding pipe for each stop.
So if you had a pipe organ with 25 stops for the "Swell" keyboard
you would have 1525 pipes!!!
The different timbres are produced by the shape, materials and
construction of the pipes themselves.
Each pipe is tuned by hand either using mechanical adjustment which
is sometimes a tab or a stopper. Since these pipes are not going to
be perfectly in tune with each other there is always some beating,
often at very low frequencies (less than 1 Hz.) Since these pipes
are located throughout a large area standing waves do not actually
stand but move as the beating occurs. This gives the sound a great
deal of movement and makes the sound appear to come from all
directions.
The Leslie speaker is an attempt to reproduce this movement.
I've often thought about what it would take to reproduce this effect
using signal processors. You can reproduce the effect to a degree
using pitch shifters set to very slight shifts and multiple
speakers, but without sending different notes through different
shifter the effect is not the same.
This is also the reason that it is very difficult to record pipe
organ (or an orchestra for that matter.) The movement of the
standing waves cannot be reproduced.
Paul Haneberg (Pipe Organ Fanatic as well as Synth Fanatic)

At 7:52 AM -0800 2/21/03, Tentochi wrote:

The responses have been VERY interesting and
insightful. But they really didn't answer my
question.

What is the relationship (if any) between the pipe
lengths marked on a synthesizer and the frequency
output. The frequencies mentioned in the replies seem
too low for what I hear on my synths.

That did me think of the psychoacoustics of pipe
organs. I don't think a lot of churches can
accomodate or afford the largest pipes. So I have
seen 3 things done to accomodate this:

1. Lay the pipe sideways on the floor.

2. Use a synthesized tone with a large subwoofer.

3. Use an implied tone. Off the top of my head, I
can't remember exactly how this is achieved. But this
is the most interesting IMO.

This also brings up the topic of resonant frequencies
of different spaces. Of of my favorite techniques
when perform is to tune my kick (TR-909) to the
resonant frequency of the performance. You want to
talk about a hell of a kick. People can't figure out
how I am doing it. I have seen a couple of other
performers also use this technique. Most of my
current work does not have any kicks in it
unfortunately. I have NOT tried pre-tuning my synths
to the resosant frequency yet. Has anyone else tried
this? I think the cool thing is that it comes in
resonating as opposed to tuning it manually during the
performance. Shake the rafters! I guess there is the
potential of literally bringing the roof down with a
sufficiently strong sound system and a sufficiently
weak roof.

Mind Your Bass Bins Mate!
Shemp

> On many traditional analog synthesizers, the octave
> was denoted by "equivalent" organ pipe lengths. I
> think the extant of the range is something like 64",
> 32", 16", 8", 4", 2", 1". Typically the range is
> some like 16" to 2".
>
> What is the corresponding frequency range for each
> of
> these pipe lengths? What notes are the pipes
> themselves tuned to? Is this the first note for
> that octave?

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

On Fri, 21 Feb 2003, Dave Trenkel wrote:

Comments in line.

Tentochi wrote:

> The responses have been VERY interesting and
> insightful. But they really didn't answer my
> question.
>
> What is the relationship (if any) between the pipe
> lengths marked on a synthesizer and the frequency
> output. The frequencies mentioned in the replies seem
> too low for what I hear on my synths.

There is no "real" relationship. The markings indicate that switching
from 16' to 8' will increase the oscillator's pitch by one octave. The
pitch is only relative. In a 1v/oct synth ths switch would merely add or
subtract 1v per "step" of this switch.

Previous posts have indicated that the pipe length is relative to the
key pressed on the organ keyboard. In most circumstances, the lowest key
on the organ manuals is two octaves below middle C. This does not mean
that the pipe played is exactly 16' if a 16' stop is drawn, but is
approximately that size. 16' feet is its "speaking" length. There are
factors of the pipe's "scale" (relative width and depth to the height)
that will have an impact on the pipe's actual length. This is more
important in an organ in that the pipe is "Hard wired" so to speak, to
the key being pressed. This is not so with an oscillator which can be
tuned to any pitch.

>
> That did me think of the psychoacoustics of pipe
> organs. I don't think a lot of churches can
> accomodate or afford the largest pipes. So I have
> seen 3 things done to accomodate this:
>
> 1. Lay the pipe sideways on the floor.
>

Pipes can also be bent (Mitered) to fold back upon themselves.

> 2. Use a synthesized tone with a large subwoofer.
>

Yawn.

I've got some Stooge cables and some from Synthesizers.com. Both are
totally fine by my standards, though I don't claim to be a cable
expert. The "dotcom" cables come in a variety of lengths, have
color-coded ends, and are reasonably priced. Just had to throw that
in because these recent cable-related posts have me wondering: What
do people have against the dotcom cables?

With apologies to other cable vendors,
John

I don't know what other people have against synthesizers.com cables, but I would hate to have to buy 4 different cable lengths just to get 4 different colors. I'd rather have 1 cable length and as many colors as I want. That's me... *shrug*

I dunno, maybe everyone should just learn to make patch cables! Then everyone could have exactly what they wanted.

On Feb 24, 2009, at 1:22 PM, Argitoth wrote:

I'm with you there - I don't have any kind of color coding going on with my cables. At one point I bought a bunch of colored heat shrink with the idea of using it for color coding, and actually used it on 2 cables, but then had the notion that it was a bit too twee for my tastes. So I've got a huge collection of all home made patch cables - all random different lengths, all black cable with mostly switchcraft style plugs but a lot of different ones thrown in there also. Different is good!

I've also got several sets of Stooge cables. I also bought a bunch
of Peavey 3' cables. They come in an assortment of colors. They're
pretty reasonable quality although three became intermittent and I
had to put a new plug on. I've also bought Calrad cables. They come
in 8 to a pack in different colors and are gold plated. I bought a
set of the 2' and a set of the 6' cables. Those have held up very
well. They also have 1/2', 1' and 3' lengths.

Dave

My cables are mostly from dot com. I have nothing against them.

http://www.synthesizers.com/cables.html
http://pugix.com/synth/about/

I have a few Stooge cables, too. Length-based color coding, like
resistor codes: brown, red, orange, yellow, green, blue for 1, 2, 3, 4,
5, 6 foot, works for me. (The 5-foot, green cables, are Stooge.) I've
had a few dot com cables short out, but they have been pretty reliable.
The plug barrels *never* come loose.

Richard Brewster

John Mahoney wrote:

there's also the GLS cables, they sell on eBay as well.
but the asst. colors in non-molded plug make these fairly nice. They aren't
the best, but they may have the best customer service I've ever encountered!
http://tinyurl.com/cablescolour

the plugs are switchcraft clones, all made in China... of course.
-p