MOTM

Wendy Carlos tinkered together a modular vocoder using modified stock Moog
modules for use on the "Clockwork Orange" soundtrack.

In her 'boxed set' notes on this instrument, bottom of page 89, she writes:

"...A special switch [on the Moog 914 Fixed Filter Bank, the filters for the
Synthesis half of the vocoder] can alter the phase of the stages, which is
the best situation for vocoding, but not neccessarily for the usual uses of
the 914..."

Does anyone know what this means?

Now, I sort of understand the importance of phase in decoding speech when
talking about the biology of the ear (the 'hairs' that get excited can
decode formants by phase-aligning to harmonics of the speech signal,
blah-blah-blah)...

...but in a vocoder, what is this all about? Squinting at the 914 photo, I
see a toggle switch that might be labeled "Norm / Alt Inv," so I'm
*guessing* that maybe it would alter the phase of every other bandpass stage
180 degrees, but... what does that do? How does that help vocoding?


On a related note, what exactly was the Roland Vocal Filter? Anybody? I
believe JH built a clone of one, but I haven't been able to find anything
about the thing at all.

> "...A special switch [on the Moog 914 Fixed Filter Bank, the filters for
the
> Synthesis half of the vocoder] can alter the phase of the stages, which is
> the best situation for vocoding, but not neccessarily for the usual uses
of
> the 914..."
>
> Does anyone know what this means?

Hmm, I thought I should know, but I really don't.
Phase relation from one band to the next is highly important in a vocoder
(and it's often overlooked by beginners who want to build their own
vocoder) - BUT the same thing is true for a filterbank.

> On a related note, what exactly was the Roland Vocal Filter? Anybody? I
> believe JH built a clone of one, but I haven't been able to find anything
> about the thing at all.

It's the filterbank for vocal formants from the Roland Vocoder Plus VP-330.
This is *not* part of the VP-330's vocoder section, but an extra filter bank
for human voice simulation. (Big surprise for me - I thought they just
had some kind of preset for the vocoder.)
There are 4 selections of human voice, male and female in different ranges,
and these are fed from different octaves and different waveforms (!)
in the VP-330. I build my clone with 4 separate inputs, each normalized
to the previous, so I can use separate input signals, or just use the first
input for all 4 options, and have a bypass switch to run the signal
unchanged.
There's also a VCA and single-trigger mode AR envelope, so you can even
get a slow attack "ahh" choir from gated (organ-like) input signals.
Of the 4 sounds, 3 are so-so, but the 8' Male is awesome - it turns any
string-like synthesizer sound into a convincing "Vangelis" choir.

I don't plan to publish the circuit - you still can get the original schemos
from Roland.

Maybe I'm selling the box at some time (I have two vocoders which can
basically do the same thing), but if I do it will not be cheap, as I spent
hundreds for an engraved custom front panel, toroidal transformer and
so on.

JH.

>> "...A special switch [on the Moog 914 Fixed Filter Bank, the filters for
the Synthesis half of the vocoder] can alter the phase of the stages, which
is the best situation for vocoding, but not neccessarily for the usual uses
of the 914..."
>>
>> Does anyone know what this means?

>Hmm, I thought I should know, but I really don't. Phase relation from one
band to the next is highly important in a vocoder (and it's often overlooked
by beginners who want to build their own vocoder) - BUT the same thing is
true for a filterbank.

The bands should be *in* phase with each other, no? Is there something good
that happens if you invert alternate bands at the output or something...?

Thanks for the info on the vocal filter. It sounds interesting! I especially
like the sound of that "Vangelis Choir"!

> The bands should be *in* phase with each other, no? Is there something
good
> that happens if you invert alternate bands at the output or something...?

They are *never* in phase. With inversion you decide for a 180deg shift,
i.e. weather you rather like your phase difference between 0 and 180deg
or 180 and 360deg. It's not like choosing "minimum difference" either. You
have to make extensive calculations (with complex numbers, i.e. amplitude
and phase) and see if you like the overall result. It's mostly a choice
between slight peaks or slight notches (or heavy peaks and deep notches,
if you started the wrong way) - don't even think of "flat" response.
OTOH, these irregularities will give a certain character to your filter bank
or vocoder - which is what we want with analogue designs - so my
_guess_ about that W.C. topic would be that she just liked more notches
for the vocoder than for the filter bank (or the other way round).
But this is really just speculation.

JH.