Tabsraction

[paranomasia]

I've had an algorithm bouncing around in my head for a few years now. It's a method of remixing songs (more specifically, entire albums). It even allows for cross genre remixing. I'm finally getting to a point that it's capable of producing (some) decent sounding cuts

Comments

$.02

[squidb0i]

Well, for beat hunk disassembly and reassembly, granular synthesis with tempo match to host would work nicely, such as the one built into Fruit7. Or more generally applicable, there are many VST plug ins that do realtime re-ordering of the audio buffer, such as dBlue Glitch, destroyFX BufferOverride, etc. These can take their timing information, from the master tempo of the host application, allowing effortless on-beat automatic buffer whacking.

I use these often. Usually the process is create some long experimental stream of sound as a file, then chop it up and use the good bits. For examples, check out some of the flow transitions in 'Last Tribe' or 'Churl' on my Myspace page..

Tonally, one would use a multi-band filter to separate the frequencies into different audio streams. While there's obviously plenty of software that will allow one to filter and process within the plug in (ie, multiband compressors, vocoders, etc) I can't think of any software that does this latter part, routing the streams to elsewhere.. hmm.. this would be worth hunting down.

At any rate, I reckon you'd want to do this part first, split the frequencies out, then play with the buffer/granular bits on a stream by stream basis. Do let us know what you come up with.. =]

Re: $.02

[paranomasia]

Yah, I started moving away from glitchy stuff when dBlue and LiveCut got popular. Their sound was too generic and it seemed like most of the people using them weren't even tweaking the parameters. That or they'd just throw it over an entire mix and basically kill anything decent that might have been there.

I'm delving into more of a 'macro-glitch' sound. Basically pushing common rhythms to ambiguity without true granular synthesis. The concept is a little goofy, but the general sound is close to Akufen. Kinda like microsampling but without the repetition of granular synthesis.

There's a command-line/terminal tool (not a plugin though) that analyzes the audio you give it, then splits it into layers based on frequency range and settings. Just a forewarning, it's NOT real time. It can take a few minutes to process a few seconds of audio. It's great for tearing loops to bits though! http://mdsp.smartelectronix.com/2005/11/simple-demixing.php

Re: $.02

[squidb0i]

oooooh, nice!
I love SmartElectronix plug ins, but was unaware of this one.
Good score, thanks!
=D