xoxos

xoxos · Post by **xoxos** » Sat Sep 06, 2008 1:10 am

:O i liked the kicks in my mass-spring demos. it is true that the only music i really listen to that uses a real kit is ~1930s, where you can barely hear it.

tbh i'm quite happy with the mass-springs.. using two instances and changing one of them is satisfying my requirements for emulative percussion at the present time, probably because they're an improvement over anything else i'm familiar with.

i'm sure they sound less convincing to others

my perception of them is coloured by my appreciation of their cpu consumption.

ZombyWoof · Post by **ZombyWoof** » Sat Sep 06, 2008 7:39 am

How about a Konnakol generator. That is, the art of spoken Indian percussion that Indian musicians use when they forget their drum. I've always loved that sound environment.

This would cross a lot of disciplines, and I haven't seen anything like it. You have quite a few building blocks already.

xoxos · Post by **xoxos** » Sat Sep 06, 2008 5:43 pm

i'll add it as a set to syng2, my current project.

it's been done before for an algorithmic application.. perhaps perry cook?? in csound or something..

ZombyWoof · Post by **ZombyWoof** » Sat Sep 06, 2008 6:16 pm

xoxos wrote:i'll add it as a set to syng2, my current project.

it's been done before for an algorithmic application.. perhaps perry cook?? in csound or something..

Nice addition!

I'll have to look for the Perry Cook experiments - thanks for the tip.

carlbe · Post by **carlbe** » Sat Sep 06, 2008 9:00 pm

ZombyWoof wrote:How about a Konnakol generator. (snip)
This would cross a lot of disciplines, and I haven't seen anything like it. You have quite a few building blocks already.

Ooooohhh - (shivers) - that would be very nice... I might be prejudice but that would win my vote for the next dev competition...
I took a quick look around and didn't find any sample libs for Konnakol...(!?!).

Go Rurik!!

ZombyWoof · Post by **ZombyWoof** » Sun Sep 07, 2008 7:44 am

Konnakol is essentially a highly sophisticated and variable human beatbox methodology. Could syng2 theoretically also include other algorithmically generated beatbox sets from other musical milieux? Can the breath control algorithm handle those percussive extremes?

xoxos · Post by **xoxos** » Sun Sep 07, 2008 4:57 pm

there is no automation of breathiness changing during a phoneme.. perhaps you could use an envelope. klatt-style synthesis is serviceable for intelligible speech but doesn't shine at the pronounciation of single syllables.

i've included a konnakol set and a bol set but that's it because it takes time to research these things. i'll probably be lynched for my nescient implementation of these two

ZombyWoof · Post by **ZombyWoof** » Sun Sep 07, 2008 5:01 pm

xoxos wrote:there is no automation of breathiness changing during a phoneme.. perhaps you could use an envelope. klatt-style synthesis is serviceable for intelligible speech but doesn't shine at the pronounciation of single syllables.

i've included a konnakol set and a bol set but that's it because it takes time to research these things. i'll probably be lynched for my nescient implementation of these two

Wow! Thanks for the effort, and the speed! Hats off, sir.

ZW

xoxos · Post by **xoxos** » Wed Sep 10, 2008 12:18 am

ZombyWoof wrote:
xoxos wrote:there is no automation of breathiness changing during a phoneme.. perhaps you could use an envelope. klatt-style synthesis is serviceable for intelligible speech but doesn't shine at the pronounciation of single syllables.

i've included a konnakol set and a bol set but that's it because it takes time to research these things. i'll probably be lynched for my nescient implementation of these two
Wow! Thanks for the effort, and the speed! Hats off, sir.

ZW

i will say "it's there, but you're going to be like 'oh.' it sounds like dectalk going 'takadimi takadimi' and lacks the technical impact of the human performance."

prepare not to be amazed when you hear a machine go 'takadimi.' :p

even if it says it twice as fast as the next machine :p trust me, even at 2,000 takadimis per second :p

thanks for the extra idea tho

i needed *some* verbal content for people to hear when demoing it.

Kaspar · Post by **Kaspar** » Mon Sep 15, 2008 5:06 pm

Hi Rurik and all!

I've only recently found your plugins.
I'm especially interested in the algorithmic ones - Circuit and Hyperion.

Now, I'd like to see a plug-in that records a MIDI loop and then makes variations to it using similar algo-ideas as in Circuit or Hyperion.

This way the initial musical idea/improvisation could be easily entered and then automatically elaborated further.
Sort of a accompaniment maker or idea generator if you will.

So, what do you think?

xoxos · Post by **xoxos** » Mon Sep 15, 2008 11:37 pm

my focus is really more on the sequence output than interactive sequencing. the performance tools fill what is imo an obvious void, i'd rather develop fully automatic music generators.

it usually takes a lot of development to fill in the blank part of that "give it some data and it spits out a whole load of good stuff."

"what do i think" i think it's pretty cool to hear that someone is intrigued by algorithmic composition.

:p

carlbe · Post by **carlbe** » Tue Sep 16, 2008 4:01 am

I'll jump in because I usually don't.

Yes - I like probilistic determinantion. There I said it.
However I like to be able to get away from the framework of BPM.
So my humble wish would be to have access to an event field that isn't organized by beats, measures, and worse of all... repeating phrase boundries.
I don't need macro - Mondays thru Thursday... but a clock-based event field would fit my work better... (smpte)and be able to cluster around clock-periods.
going back to lurksville... cheers

Kaspar · Post by **Kaspar** » Tue Sep 16, 2008 5:50 am

xoxos wrote:my focus is really more on the sequence output than interactive sequencing. the performance tools fill what is imo an obvious void, i'd rather develop fully automatic music generators.

it usually takes a lot of development to fill in the blank part of that "give it some data and it spits out a whole load of good stuff."

OK, fair enough.
I was just hoping somebody would take on the hard work of teaching machines to guess what the composer would like to hear

I'll continue to investigate if Circuit and Hyperion can meet my needs and post to the general forum if I'm in trouble.

Thanks!

xoxos · Post by **xoxos** » Tue Sep 16, 2008 6:00 am

carlbe wrote:I'll jump in because I usually don't.

Yes - I like probilistic determinantion. There I said it.
However I like to be able to get away from the framework of BPM.
So my humble wish would be to have access to an event field that isn't organized by beats, measures, and worse of all... repeating phrase boundries.
I don't need macro - Mondays thru Thursday... but a clock-based event field would fit my work better... (smpte)and be able to cluster around clock-periods.
going back to lurksville... cheers

well, there you have it

http://www.xoxos.net/sem/sem.html

i updated the page recently to include some bits and pieces i've been using in the sequencers.

i don't want to be anyone's font (ergo i'm not the greatest businessman..) and i have mentioned an agenda of increasing resources for algorithmic development in synthedit. i haven't had the time to put into the project more than (finally) updating the page and the video a while ago, and tbh i really never hear from anyone about algorithmic development in synthedit.

almost a decade into predominantly computer-based music and algorithmic music is still not recognised, let alone integrated with pop culture. to me, it's "the next logical step in composition.." synthedit has been suitable for me.

being perhaps the entity to-be of "bringing algorithmic pop music to the masses," my musical sensibility is defined by clocks and measures.. which is probably why i default to 'dense uniform clocks' in my recordings :p

that's my thing - what's yours?

carlbe · Post by **carlbe** » Wed Sep 17, 2008 3:59 am

sorry for the slow uptake

- banging my head on that same wall again. Dumb monkey.

Fortunately I have no time to invest in synthedit thinking that more kindred process folks are found in PD or MAX, Usine, even Nord G2, so energy goes there, but thanks for the very generous non-null pointer.

I hope you algo-pop the planet massively, I do, and I did enjoy your latest pieces via synging and other modeled instruments. They were really great. Solar System gives me that Entergalatic Sound of Joy feeling. Your models are nice from the first note; immediacy. The bass in syng2pop is very expressive - killer pitch management. Organic. Nuff ass-kissing, cramps.

I was taken with your "naturalistic" models, having been intro'd by your winning the dev contest (who says they don't help create exposure?

) and thought that the creation of soundscapes using them was neato. (e.g. the "rains" mp3 on breathcube) One can go into nature but those little dickens just don't cue properly - except mosquitoes - they're dependable. But smpte based organization is more useful than BPM for me. Not going to see a weighted probability clustered set of squeakers around the 3:12:29 mark on the Café eh? My best bet is to stop pestering you and wrap fauna etc. in Usine or Max.

Ok - please have this post deleted in the spirit of font evaporation (ugh puns)! It was more fun to write than reread.

sunship

xoxos

KVR Audio

request new plugs