vocoder dim

[Martin Czech]

To all vocoder experts :

I'm just designing my 20-Channel Vocoder 
(only Spice-Simulation up to now).

I'll use 10th order Bessel bandpas-filters (analysis),
60dB/oct (my tables end at 10th order ;) )
with 10 op-amps and 20 caps.
The overall impulse response will always show some
"ringing" since the vocoder analysis filters are a bank of resonators.
I don't think that there is a way to
avoid this. Maybe with digital filters, but that off topic here.
Ok, the Bessel has not so sharp edges,
but less "ringing" than Butterworth or Chebychev, 
so I think this will be optimal.

Does this make an audible difference ? Do you agree ?

The synthesis filter bank will be the same  and I'll provide
jacks so that both banks can be used independently as
fixed filter bank. Indeed, every component of the vocoder
should be patched, no pre-wiring.

So it'll be also possible to bypass some bandpass-filter
stages to get not so steep responses (r.g. 6bB, 12dB, 24dB & 48dB & 60dB).
This will also give "cheaper" vocoder sounds.

I'm currently thinking about an bargraph led display of the 
analysis input (one for each channel) with National
bargraph driver ics.

After rectifying the filtered analysis input the signal must be low
pas filtered to get the synthesis control voltage.
What kind of lp filter is recommended and what
is the best cutoff frequency relative to bandpas 
frequency ?

My simulations show that a 4th order Tchebychev lowpas
with cutoff = bandpas cutoff gives good ac ripple suppression
while not beeing tooooo sloooow.
Ripple suppression might get important if the analysis and
synth section are not connected in the usual way but 
for example inverted (lowest band to highest band etc).
In this case any control voltage ripple should give non
masked audible am-modulation artefacts.

Some EMS Vocoders have slew regulation for rising and falling
edge of the analysis detector simultaneously with a single pot
for all channels. Interesting feeeaaatuuuure.
How ?
Voltage controlled RC slew regulation with junction fet as "R" ?
They also have a freeze feature. 
Could also be done with "RC"-fet as sample & hold.
Does anybody know if the EMS freeze function suffers from leakage
after some time ? That would be a hint.

The Formant Vocoder uses a highpas-filter for the voiced/unvoiced
detector. I think that any filtering for this purpose is unnecesary
since the analysis filter bank already gives spectral information 
with good resolution. Combining the filter bank  envelope outputs
in some way (analog fuzzy logic) should give the desired voiced/unvoiced
information not only as digital "1" and "0" but in an analog way :
voiced -> more hiss -> much more hiss -> totally unvoiced.
A simple vca could add more and more white noise to the synthesis
filter bank according to the fuzzy signal.

Any information would be helpfull.


m.c.