SirSickSik Contributions

#683

NEW MODULE

HARMONY

"ribbonKeyboard"
a 32-note keyboard mostly for use when developing a synth and you want a quick keyboard.
white and black notes are shown on the bar. When clicking a note, it only shortly flashes and returns to former state immediatly after.
Black notes aren't triggered, only the white ones, so you can also "swipe" the ribbon and play an arpeggio.
NOTE: not really useful for live-performance use (pretty pointless) as gate-triggers are also re-send when you move your mouse a bit but still stay at the same note.. This can be partly overcome to add a pulse-timer module to the trg output

It's use for development is nice AND you'll have a visual feedback on the scales of my scalebank module-> when changing scale/key, it updates the keyboard to the selected scaling/key,

#684

NEW MODULE

MIDI

"QuneoPoly4" and "QuneoPoly4Rec"
4x poly midi2cv keyboard module for use with the QUNEO midicontroller, with for each voice independent outputs:
gate
note
velocity
pressure
CV1 (X-axis)
CV2 (Y-axis)

The recorder can record 4 voices into an array. With a fast counter (count input) you can record live automation from your quneo!

quneo-layout is explained in the "overview" of the module

#686

sweet, managed to combine a new vowel controller+oscillator+filter to be controlled using the text2index module :stuck_out_tongue: Most of the vowels+variations and some soft consonants are created, now a good system for creating the noisy/harsh consonants. Lowered the cpu use of the controller quite a bit in comparison to the old one, though now I'm using 5 channels of 3rd order biquads to do the filtering, though it asks a bit of cpu, it sounds soooo sweet :stuck_out_tongue:

#687

sounds really interesting !...

#688

keep us posted when you release it!!! sounds good.

#689

Does sound interesting! I'm a bit obsessed with vintage speech-synthesis methods, so very much intrigued to hear how this sounds.

Keep up the great work!

a|x

#690

oh, I must say I'm going the experimental route. I have seen a couple of papers about the subject, but it's often quite complex material with lots of formula's and for-me unknown symbols. So I'm just listening very closely to my own consonents and try to figure out how to remake these.. eg. the R and Z will be done by a sine AM (or sine-saw-shaped?) over de vosim/noise input respectively before it's going into the vowel filters (or after, still gotta work out the consonant-oscillator and then gonna test what works best).
The most difficult ones will be D and B (short basskick-like sines?)
Most important thing is to make it "understandable" first so I can check the "smoothness" of the voice crossing from vowel to consonant and back, whether the consonant seems to be the same voice or sounds like it's an external click/buzz.

#691

That's cool.
There are lots of ways of generating speech-like sounds. Stringing them together into comprehensible speech is the really complicated bit, I think. You could have a look at the reverse-engineered SAM speech-synth from the Commodore 64 if you're interested in rule-based text-to-speech methods.

a|x

#692

I want to do this fully in synthesis, so no use of prerecorded samples or wavetables, just trying to use bare synthesis that could be done using basic modules (oscs, VCA's and filters) on a modular.
So in the end these modules should also provide a layout how to work this out on your modular with basic modules and use the axoloti to only provide the right settings.
I think lots of the available speech synthesizers have used wavetables/waveguides to provide sounds for consonants, at least, that's what I've mostly found out till now and is something I don't want to do. I haven't been able to find something that goes from basic synthesis-styles like FM or AM into consonant speech synthesis.

As for the sequencing part, the vowels are already very nicely morphing into each other (done in the same way I morph my wavetable oscillators, also allowing a curve to be superimposed on the morph between vowels), so now it's all about placing the consonants to mix in filtered noise and modulations to turn them into the needed consonants when needed.

#693

I hope that people don't mind, but I've added yet another function to my rndPattF module.
These are controls for setting the chance and change size of non-repeating random offsets from the (random/recorded) pattern being played.

and just remembered about another one I wanted to add...
a random "non-sample" chance, disabling the output to be updated for a maximum random amount of steps. (*fixed, and yes, this is very very nice indeed :P)

#694

NEW MODULE

DIST

"signalsync"
A nice take on syncing signals.
Two inputs: sync signal and an audio signal to be synced to that signal.
Audio signal is put into a buffer and last zero-crossing in the buffer is reminded.
Once the sync input goes from negative to positive, the buffer readout is adjusted so it restarts from the zero-crossing.
The module does use an overlap to prevent harsh harmonics. As soon the the sync goes positive, the last buffer being played is faded out while the new buffer is faded in. Even if the next sync comes quicker than the fade out of the former buffer, the offset that would remain is also faded out.

useful for syncing signals for AM-synthesis that don't have a hard-sync

#695

gltchbeast!! This is really great!!! Thanks a lot, you are supergreat man!
my only problem:
When MIX is set to max on the module,
the DRY signal is still coming through when using Grepeater...

#697

that behavior was intended. I wanted a repeating delay that added to the incoming sound.
If you don't want the dry signal , you need to do this:

-embed module
-go to S-rate Code
-search for line 91 and 92.
-behind this I've just added an extra code in green.
-now just remove everything on the lines before the green code (also the green //OR// part).

this code subtracts the dry signal while adding the repeater as soon as the volume of the Grepeater goes up.

#698

Hi SirSickSik, thanks that looks great.
Unfortunately I cannot find the new extra green code.
I've synced my library several times, even removed the glitchBeast from my folder and let it load in again, just to make sure.
But even in this fresh version I cannot see it either...
Thanks for your help!

#699

I've just hit "sync libraries" an extra time, maybe I forgot to do that ..

#700

Yes! It's working!!!
But actually it sounds better the way it was. It seems the RepeatSize does not behave properly anymore, or at least doesn't sound as good as the original version.
I thought it would behave more like an endless 'buffer repeat' when 100% wet, but I guess this works a bit different regarding feedback maybe?
Another thing I'm still struggling with on the glitchBeast is to get it temposync (slave)...
I'm looking at your Clock2Timing module, but I don't find an input on the glitchBeast.
Where should I connect it?

Thanks again and greetings!

#701

the inputs of the timing-parameters respond the normal way like the factory timings.
So you can use the delaybase output of the clock2timing module combined with one of the ratio-modules (I have several different modules to quickly set integer ratio's->sss/math) and connect these to the timing-parameters.
Same counts for the LFO rate's, though remember, frequency behaves inversely from delaytime.
Multiplying a frequency makes it faster, thus a shorter cycle, while multiplying time makes it longer, thus a longer cycle,

For buffer repeat, there's the "Gretrig", combined with the gapper this functions as a gapping repeater. Note though, that this retriggering buffer goes through the whole fx-system, thus updating the buffer with new material every loop.

#702

@jaffasplaffa
Thanks for the heads up, and sorry to hear you had to return the compressor kit!

@SirSickSik
Thanks for updating the module, I should be able to make use of that with a bit of luck!
As for the huge multi-touch screen, dude, you're gonna look like an octopuss when using it :grin:

#703

playing around with my new "preset-morphing" module, holy shit, this is awesome XD
74 parameters of drum modules being morphed between 64 presets. Thanks to finally figuring out how to get the values of parameters in other modules :slight_smile:
Just using note information to set preset number and velocity/aftertouch for morph-rate and whether to "get" the (randomised) settings of the modules to a preset or to morph to another preset.

#704

looking forward to implementing it when I properly get my head around axoloti coding