This seems to require an analysis step, alhtough it reads more
like subtractive synthesis than GS. There is however an analogy to
FOF (et al): what you get from FOF is equivalent to a
pulse train delivered to an IIR filter; each grain is a filter impulse
response (which overlaps with others), starting at every T secs,
for a fund. of 1/T. The fofilter opcode simulates this filter, so
if it gets an impulse train as input, it is quite similar.

Victor
----- Original Message ----- 
From: "Tim Mortimer" 
To: 
Sent: Friday, January 11, 2008 2:04 AM
Subject: [Csnd] Re: Britney Spears of Destiny (esp. attn Victor Lazzarini + 
others)



Thanks everyone!

Very quickly (while I bludge off work) here are my responses to you all…

Victor:

Yes I’ve identified all those opcodes you have suggested as being “close to
the money” but as far as implementing what Roads is talking about in Com Mus
Tut, I’m not sure any of them deal with, well, pretty much any of it. Please
correct me if I’m wrong.

These are the key quotes from CMT (p174)

"spectrum analysis is performed on each grain?" (in a window that's == grain
size?) (With grains == successive wavecycle lengths?)

"an impulse response is then derived which sets the params on a resynth
filter..."

"pulse trains delivered into FIR filters at the detected pitch period.."

"The output signal results from the excitation of the pulse train on the
weighted sum of the impulse responses of all the filters"

This seems like a specific mix of a number of different DSP philosophies &
approaches – no simple “Foggery” or “Synciness” here…. But as you are across
so many of these related topics, you had the misfortune of having me drag u
in…. (I apologise ; ) …)

It's time i grappled further with the intricacies of filters & impulses i
think. (beyond a basic understanding of convolution, which is about where
i'm at at he moment...)

I DO see the analogy you draw below however ... I'm off to explore what i
can of this further in Computer Music Tutorial...

Tried Melodyne demo - it's a one trick pony maybe - but boy, what a trick!
(& probably a lot more versatile & useful i'd hazard than perhaps what many
of you are probably thinking.... does a VERY good job of repitching &
timestretching vocals, brass, even percussion (the latter sans formant
tweaking obviously... hardly any artefacts at all even under very extreme
settings & the interfacing is pretty good too!)

I hope Victor (& any other developers watching) have drawn something of
worth out of this - whether i prove capable of executing this using csound
or not i've manage to elucidate (in a round about way) many of my wishes &
desires for some future areas for csound development i hope... & uncovered
further areas where i myself need to learn more & experiment further with
Csound...


Victor.Lazzarini wrote:
> 
> This seems to require an analysis step, alhtough it reads more
> like subtractive synthesis than GS. There is however an analogy to
> FOF (et al): what you get from FOF is equivalent to a
> pulse train delivered to an IIR filter; each grain is a filter impulse
> response (which overlaps with others), starting at every T secs,
> for a fund. of 1/T. The fofilter opcode simulates this filter, so
> if it gets an impulse train as input, it is quite similar.
> 
> Victor
> ----- Original Message ----- 
> From: "Tim Mortimer" 
> To: 
> Sent: Friday, January 11, 2008 2:04 AM
> Subject: [Csnd] Re: Britney Spears of Destiny (esp. attn Victor Lazzarini
> + 
> others)
> 
> 
> 
> Thanks everyone!
> 
> Very quickly (while I bludge off work) here are my responses to you all…
> 
> Victor:
> 
> Yes I’ve identified all those opcodes you have suggested as being “close
> to
> the money” but as far as implementing what Roads is talking about in Com
> Mus
> Tut, I’m not sure any of them deal with, well, pretty much any of it.
> Please
> correct me if I’m wrong.
> 
> These are the key quotes from CMT (p174)
> 
> "spectrum analysis is performed on each grain?" (in a window that's ==
> grain
> size?) (With grains == successive wavecycle lengths?)
> 
> "an impulse response is then derived which sets the params on a resynth
> filter..."
> 
> "pulse trains delivered into FIR filters at the detected pitch period.."
> 
> "The output signal results from the excitation of the pulse train on the
> weighted sum of the impulse responses of all the filters"
> 
> This seems like a specific mix of a number of different DSP philosophies &
> approaches – no simple “Foggery” or “Synciness” here…. But as you are
> across
> so many of these related topics, you had the misfortune of having me drag
> u
> in…. (I apologise ; ) …)
> 
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 
> 

-- 
View this message in context: http://www.nabble.com/Britney-Spears-of-Destiny-%28esp.-attn-Victor-Lazzarini-%2B-others%29-tp14730945p14753149.html
Sent from the Csound - General mailing list archive at Nabble.com.