Hello list, After a few years of maxing I migrated back to PD about a year and a half ago and havent looked back. Now I ran into a stumbling block and I figured I turn to you for help:
Im looking for a way to impose a harmonic spectrum on a non-harmonic signal. I tried cross-synthesis with a harmonic signal, which is ok, but what Im actually trying to do is to tune a field recording, in a way, so it can have harmonic relations to a live instrument.
I played with the pvtune~ object from fftease which is nice, but the results end up sounding a bit autotune-ish. And I also tried driving a few vcf~ filters with frequencies running from fiddle~/sigmund~ but the fundamental from these objects is too erratic and it sounds very unnatural.
Wishart describes in his book a technique he calls spectral focusing, which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but Im not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Any help with this would be greatly appreciated, Zax
Hi Zax,
Maybe a straight up vocoder is too primitive for the kind of sound you're after, but who knows... For what it's worth, I have a vocoder patch in my timbreID examples package with filters that are spaced evenly in Barks. You can also tune the relative amplitudes of the filter outputs.
Of course, it still might just make everything sound like a robot :)
On Wed, May 25, 2011 at 9:24 PM, eran@mx.kein.org wrote:
Hello list, After a few years of maxing I migrated back to PD about a year and a half ago and haven’t looked back. Now I ran into a stumbling block and I figured I turn to you for help:
I’m looking for a way to impose a harmonic spectrum on a non-harmonic signal. I tried cross-synthesis with a harmonic signal, which is ok, but what I’m actually trying to do is to “tune” a field recording, in a way, so it can have harmonic relations to a live instrument.
I played with the pvtune~ object from fftease which is nice, but the results end up sounding a bit autotune-ish. And I also tried driving a few vcf~ filters with frequencies running from fiddle~/sigmund~ but the fundamental from these objects is too erratic and it sounds very unnatural.
Wishart describes in his book a technique he calls spectral focusing, which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but I’m not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Any help with this would be greatly appreciated, Zax
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
@ William Brent: William, I'm a big fan of tID - what an incredible library! I've used it for several projects this year. So first up - big thanks. I was able to get some interesting results from the bark-vocoder(not at all robot like!), with replacing the [osc~]s with [vfc~]s and tracking the fundamental of a harmonic source with [sigmund~]. But this still sound like imposing the harmonic characteristic on an existing spectrum, while I think I am looking to arrive at harmonicity through something like moving the frequencies of the partials into their closest values according to the harmonic series (so moving from bins=>partials).
Can this be done? it would entail changing the frequency of each bin separately. A related procedure would be non-linear spectral stretching, which I also could not find in my searches through pd-world.
@ Peter Plessas: Miller's timbre-stamp is totally cool, but if I understand it correctly it changes the spectral contour of the sound by replacing the amplitudes of the respective bins, but as is the case with fft, the frequencies of the bins are determined by the analysis-size so they remain the same for both sounds. And while this sounds great I think I'm looking to change the frequency of each partial in a way that isn't tied with the discrete frequencies of the bins..
This is what got me curious about Wishart's way of using lpc - he says that you can set the coefficients according to pitch relations, though I still dunno how to do this in pd, argh...
Thanks guys, Zax
Hi Zax,
Maybe a straight up vocoder is too primitive for the kind of sound you're after, but who knows... For what it's worth, I have a vocoder patch in my timbreID examples package with filters that are spaced evenly in Barks. You can also tune the relative amplitudes of the filter outputs.
Of course, it still might just make everything sound like a robot :)
On Wed, May 25, 2011 at 9:24 PM, eran@mx.kein.org wrote:
Hello list, After a few years of maxing I migrated back to PD about a year and a half ago and havent looked back. Now I ran into a stumbling block and I figured I turn to you for help:
Im looking for a way to impose a harmonic spectrum on a non-harmonic signal. I tried cross-synthesis with a harmonic signal, which is ok, but what Im actually trying to do is to tune a field recording, in a way, so it can have harmonic relations to a live instrument.
I played with the pvtune~ object from fftease which is nice, but the results end up sounding a bit autotune-ish. And I also tried driving a few vcf~ filters with frequencies running from fiddle~/sigmund~ but the fundamental from these objects is too erratic and it sounds very unnatural.
Wishart describes in his book a technique he calls spectral focusing, which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but Im not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Any help with this would be greatly appreciated, Zax
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- William Brent www.williambrent.com
Great minds flock together Conflations: conversational idiom for the 21st century
www.conflations.com
I think I am looking to arrive at harmonicity through something like moving the frequencies of the partials into their closest values according to the harmonic series
A while ago I started an extern that does things in the same ballpark as you're describing...maybe I can add that specific kind of processing as an option as well. If you're not up against a deadline I can let you know when I get a chance to experiment with it in the next few weeks.
Wishart describes in his book a technique he calls spectral focusing,
which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but I’m not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Ah...LPC in PD.
I had a go at turning another speech processing system into a PD external with a view to being able to manipulate spectral envelopes and components about 5 years ago, but with no success. What LPC does is split the signal into filter coefficients and excitation - so that the excitation passed through the filter will reconstruct the original signal.
There is still no LPC plugin that really works in PD, and my attempt was glitchy and horrible. What would be better would be a pitch+noise separation process, such as is used by Spectral Modelling Synthesis by Xavier Serra at UPF in Barcelona. It is still available as part of the CLAM package ( http://clam-project.org/%C2%A0).
Perhaps the iem_spec library contains some useful FFT related stuff, but to be able to manipulate each frequency component individually is a formidable challenge!
Best, Ed
Hi Zax, did you look into Miller's "timre stamp" example file? Perhaps this is also useful for you! best, P
eran@mx.kein.org wrote:
Hello list, After a few years of maxing I migrated back to PD about a year and a half ago and haven’t looked back. Now I ran into a stumbling block and I figured I turn to you for help:
I’m looking for a way to impose a harmonic spectrum on a non-harmonic signal. I tried cross-synthesis with a harmonic signal, which is ok, but what I’m actually trying to do is to “tune” a field recording, in a way, so it can have harmonic relations to a live instrument.
I played with the pvtune~ object from fftease which is nice, but the results end up sounding a bit autotune-ish. And I also tried driving a few vcf~ filters with frequencies running from fiddle~/sigmund~ but the fundamental from these objects is too erratic and it sounds very unnatural.
Wishart describes in his book a technique he calls spectral focusing, which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but I’m not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Any help with this would be greatly appreciated, Zax
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Or convolution?
Hi Zax, did you look into Miller's "timre stamp" example file? Perhaps this is also useful for you! best, P
eran@mx.kein.org wrote: Hello list, After a few years of maxing I migrated back to PD about a year and a half ago and haven’t looked back. Now I ran into a stumbling block and I figured I turn to you for help:
I’m looking for a way to impose a harmonic spectrum on a non-harmonic signal. I tried cross-synthesis with a harmonic signal, which is ok, but what I’m actually trying to do is to “tune” a field recording, in a way, so it can have harmonic relations to a live instrument.
I played with the pvtune~ object from fftease which is nice, but the results end up sounding a bit autotune-ish. And I also tried driving a few vcf~ filters with frequencies running from fiddle~/sigmund~ but the fundamental from these objects is too erratic and it sounds very unnatural.
Wishart describes in his book a technique he calls spectral focusing, which relies on lpc. He says that lpc can be set to generate a filter bank which is distributed evenly with respect to pitch, but I’m not sure I understand what he means and how it can be done in PD. lpc seems to output a set of amplitudes for the coefficients. Can I set the filters to frequencies of my choice?
Any help with this would be greatly appreciated, Zax
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Ed Kelly -
eran＠mx.kein.org -
Peter Plessas -
William Brent