![]() these are the kind of things that just take a good day of getting your hands dirty (after spending a good week reading about the process and not getting it day after day, and feeling like a moron!). ![]() Pitch shifting is achieved with a dual head crossfading delay (ala Eltro Tempophon/Dennis Gabor/Pierre Schaeffer phonogene) and is. pitch shifting, audio editing, formant shifting, etc. SoundHack Delay Bundle is a group of four plugins derived from the same basic delay algorithm: a hermite interpolated delay line with variable modulation, and a feedback loop with dc blocking and saturation. I can't afford a Pacarana, so until then, I process my cross-modulations offline in max.ĭo it by hand. Not because they're idiots or anything - it's just not very straightforward unless you understand the basic concepts behind it (Fourier Transform). Other than KSP scripting your own instruments from scratch, this is as close to Voodoo and black magic you can get in Kontakt - it would take a good few pages to walk you through the process, and most people just don't understand what their doing. you can use an AET filter, analyze a source sound, and apply the phase characteristics to another sound. lol fancy phrase for short term Fourier transform). It has a processing filter in it called an "AET Morph". The last option, which just sprang to mind, is Kontakt. If you're unfamiliar with DSP, and you don't have the cash laying around for a Kyma Pacarana, then chances are you'll have to do this by hand, with a LOT of resampling. Spectral convolution would work if you played with it enough, but the reality is you're looking at a lot of resampling, using a convolution reverb on the source sound - with the other sound loaded into a good convolution reverb program. That's what happens when you spend 3k+ on a box with like 16 multicore CPUs in it dedicated to audio though. Essentially the same thing - but it's in real-time, and incredibly high-detail. If you've got a few grand to spend, get a Pacarana from Symbolic sounds - and you'll have access to what they refer to as "Tau Morphing". ++delay a fully featured 10 second delay, with extensive modulation and feedback filtering. This provides a positive or negative frequency shift that is somewhat complicated by the wrapped bands. ++pitchdelay for old school multitap delay-based pitch-shifting. ++bubbler an extremely flexible and MIDI playable granular delay. then it's just a matter of loading the two samples together and using a UI object (like a slider) to morph them together. SoundHack Delay Kit includes: ++flipper, for synchronized reverse delay. The formant structure can be regarded as the filter that defines the character of a voice (particularly the vowels) or a musical instrument. You don't need to necessarily know the hard math behind it, but understand what it does, and how to properly utilize windowing and bins in the context of MSP to get the sound you're after. you can either create a patch in max 6/7 using the pfft~ object, which can get a little tricky, seeing as you'll need to learn about the Fourier Transform. If that's indeed the case then there are a few ways to go about doing this. 65-71.The "morph" you're describing is known as (both) spectral cross-modulation, and spectral convolution.įrom what i've read in your question, I gather you're interested in essentially crossfading the two sounds together, but in the frequency domain not the time domain? Keith Lent, "An Efficient Method for Pitch Shifting Digitally Sampled Sounds", Computer Music Journal Vol.It is distributed as a Quark for SuperCollider. This pseudo-UGen was created by Marcin Pączkowski at DXARTS at the University of Washington, and is based on Juan Pampin's and Joseph Anderson's implementation of Keith Lent's pitch shifting algorithm. The technique used is also known as Pitch Synchronous Overlap-Add synthesis (PSOLA) or Pitch-Synchronous Granular Synthesis (PSGS). Good quality of tracking is crucial for achieving good results. PitchShiftPA works only for single-voiced (monophonic) sounds and requires tracking of their fundamental frequency (e.g. shifting pitch while maintaining the formant structure, or changing the formant without changing the pitch. ![]() Phase Aligned Wavepacket (Re-)Synthesis (PAWS) pseudo-UGen, allowing for manipulating pitch and formant structure of the sound independently.
0 Comments
Leave a Reply. |