Yeah, software tied to specific hardware sucks. In a world where you get paid according to how much money you get away with charging, rather than for the actual worth of your contribution to society, not all companies are nice or wise enough to avoid that kind of lock-in.
Still, good to explore different options, and you never know you may end up with some awesome sound by combining tools that you wouldn't have looked at if you could afford it.
XSynth, by the way, is not only a reason plugin; there was a totally free vst of that name kicking around, though a quick search has failed to find it... then again you'd probably get similar results with ableton vocoder and resampling... good luck!
How To Do Spectral Morphing
Re: How To Do Spectral Morphing
Ableton vocodoer huh! I'll look in to that!
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
I bought Max For Live last night. Monolake has created several instruments for Max For Live. One of them is Granulator. This is so far, the closest thing I've heard to what I want. I was able to make a useful instrument out of a bagpipe sample straight away. That's more than I can say for the other instruments I've looked at so far (including Alchemy). Another patch in Reaktor called Mirage by Peter Dines was similar. Although I didn't have as much luck with that in terms of creating useful sounds but that could just be my lack of ability.
Does anyone recommend any other Max or Reaktor patches for this kind of thing?
Is there a good convolution reverb for Max or Reaktor?
PS: I tried the Ableton vocoder. I can't really understand how it works. How can a Vocoder work as an effect? My understanding is that a Vocoder always requires a midi input. So, it needs to be an instrument. I guess I'll have to read the documentation about the Vocoder. Anyway, I often use Reaktor Razor's vocoding. It has some killer vocoders but unfortunately, it's still not the sound I'm looking for. It kills too much of the organic sound.
PS: Just now listening to Monolake's music. Monolake's music is another example of the sound I was originally mentioning. Monolake also takes organic sounds from the environment and then turns them in to something musical. I'll be looking at all of his instruments and seeing what I can learn from this guy.
Does anyone recommend any other Max or Reaktor patches for this kind of thing?
Is there a good convolution reverb for Max or Reaktor?
PS: I tried the Ableton vocoder. I can't really understand how it works. How can a Vocoder work as an effect? My understanding is that a Vocoder always requires a midi input. So, it needs to be an instrument. I guess I'll have to read the documentation about the Vocoder. Anyway, I often use Reaktor Razor's vocoding. It has some killer vocoders but unfortunately, it's still not the sound I'm looking for. It kills too much of the organic sound.
PS: Just now listening to Monolake's music. Monolake's music is another example of the sound I was originally mentioning. Monolake also takes organic sounds from the environment and then turns them in to something musical. I'll be looking at all of his instruments and seeing what I can learn from this guy.
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
hi kruddler.
i stumbled upon this topic. it has been fun to read of your journey.
indeed, if you are now m4l-ed up, looking at all of the henke devices is definitely a good place to start. a lot of these are granular, not spectral, of course. make sure you download them from his website as those ones are more up to date then the ones at ableton site / maxforlive.com site (i think?!).
in case you are still into the specifically spectral side: max has been doing fft (spectral analysis) very well for years and years. there are many examples floating around. specifically for m4l i would shamelessly plug one of my 'jg.delays' devices here:
http://www.maxforlive.com/library/device.php?id=158
not 'morphing' but interesting i hope.
also, this thread has talked about spectral resynthesis. there is a great object in msp called 'ioscbank~' which can help build large arrays of sines if you feel like digging deeper into the guts. a great example of this (not its use in spectral stuff, but as a basic oscillator use) can be found in the typically quirky and inventive and brilliant katsuhiro chiba's work, his synth 'dotsch'. scroll a little down this page to download (it is max, not m4l, but completely open):
http://www.audiooo.com/jp/software/index.html
hth
(i'll post back if i think of more specific spectral things).
i stumbled upon this topic. it has been fun to read of your journey.
indeed, if you are now m4l-ed up, looking at all of the henke devices is definitely a good place to start. a lot of these are granular, not spectral, of course. make sure you download them from his website as those ones are more up to date then the ones at ableton site / maxforlive.com site (i think?!).
in case you are still into the specifically spectral side: max has been doing fft (spectral analysis) very well for years and years. there are many examples floating around. specifically for m4l i would shamelessly plug one of my 'jg.delays' devices here:
http://www.maxforlive.com/library/device.php?id=158
not 'morphing' but interesting i hope.
also, this thread has talked about spectral resynthesis. there is a great object in msp called 'ioscbank~' which can help build large arrays of sines if you feel like digging deeper into the guts. a great example of this (not its use in spectral stuff, but as a basic oscillator use) can be found in the typically quirky and inventive and brilliant katsuhiro chiba's work, his synth 'dotsch'. scroll a little down this page to download (it is max, not m4l, but completely open):
http://www.audiooo.com/jp/software/index.html
hth
(i'll post back if i think of more specific spectral things).
Re: How To Do Spectral Morphing
Glad you found something that does what you wanted 
Vocoders:
A vocoder is an effect which applies the spectral shape (formant?) and rhythm of one stream of audio, to another stream of audio. It does this by chopping up the second stream (the carrier) into lots of frequency bands and adjusting the gain of each band to match the amount of that band present in the first signal (the modulator). There are various ways of doing this but all are in the same family.
Just to confuse you, some vocoders do behave like synths with a midi input. In this case the carrier is not an external signal but is synthesized inside, either as an oscillator, or white noise, or sample etc. I think the ableton one has options for both internal/external carrier.
But obviously you can take a vocoder with external carrier and plug it onto the end of a synth which you play normally. Then use whatever (drum track, mic etc) as your modulator input.
Vocoders:
A vocoder is an effect which applies the spectral shape (formant?) and rhythm of one stream of audio, to another stream of audio. It does this by chopping up the second stream (the carrier) into lots of frequency bands and adjusting the gain of each band to match the amount of that band present in the first signal (the modulator). There are various ways of doing this but all are in the same family.
Just to confuse you, some vocoders do behave like synths with a midi input. In this case the carrier is not an external signal but is synthesized inside, either as an oscillator, or white noise, or sample etc. I think the ableton one has options for both internal/external carrier.
But obviously you can take a vocoder with external carrier and plug it onto the end of a synth which you play normally. Then use whatever (drum track, mic etc) as your modulator input.
-
Press Enter
- Posts: 205
- Joined: Mon Aug 22, 2011 4:06 pm
Re: How To Do Spectral Morphing
the "pacarana" dsp is a whole, whole, whole lot cheaper & more powerful than it's predecessor. And if you gotta get into some mad cost comparison buzz, it's in the same ballpark as a polyphonic analog synth. I'd reign in the ranting and take stock.
Re: How To Do Spectral Morphing
ciw,
Thanks heaps for the tip on Vocoding. I finally had a play with Ableton's Vocoder. Let me just say, it's fucking great. The only exposure I'd had to vocoding was in a couple of different instruments and in Logic Pro. Basically they just allow you to mix something with a rhythmic quality with the instrument's own subtractive synth. This only ever creates the classic robot voice shit that noone wants to listen to these days. But, Ableton's Vocoder opens things up completely. It essentially allows you to mix anything with a rhythmic quality with anything with harmonics.
Here's a brief rundown of how to do it:
-Create two tracks.
-Put a drum beat in the first track
-Put a melodic instrument in the second (choose something with a long sustain to start with)
-Create a midi clip in the second that plays a note for a while bar
-Put the vocoder effect on the first track
-Set the vocoder to external
-Press play
You'll hear the rhythmic quality coming through melodic instrument. For example, I was able to create a piano sound that jangles (for lack of a better word) at the end. You can create this effect by putting beat repeat on the drum beat.
This has really opened up a lot interesting and useful sounds to me. I didn't realise that it would be possible to use Vocoding in this way. Thanks.
Thanks heaps for the tip on Vocoding. I finally had a play with Ableton's Vocoder. Let me just say, it's fucking great. The only exposure I'd had to vocoding was in a couple of different instruments and in Logic Pro. Basically they just allow you to mix something with a rhythmic quality with the instrument's own subtractive synth. This only ever creates the classic robot voice shit that noone wants to listen to these days. But, Ableton's Vocoder opens things up completely. It essentially allows you to mix anything with a rhythmic quality with anything with harmonics.
Here's a brief rundown of how to do it:
-Create two tracks.
-Put a drum beat in the first track
-Put a melodic instrument in the second (choose something with a long sustain to start with)
-Create a midi clip in the second that plays a note for a while bar
-Put the vocoder effect on the first track
-Set the vocoder to external
-Press play
You'll hear the rhythmic quality coming through melodic instrument. For example, I was able to create a piano sound that jangles (for lack of a better word) at the end. You can create this effect by putting beat repeat on the drum beat.
This has really opened up a lot interesting and useful sounds to me. I didn't realise that it would be possible to use Vocoding in this way. Thanks.
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
No worries, glad to have been of use
Re: How To Do Spectral Morphing
This has turned in to more of a blog about my experience than an actual forum thread. Here's some new info. I discovered that Max has a Spectral Harmonics engine and I have been able to transpose notes with spectral harmonics as opposed to simple pitch shifting. This a big step forward and I'll be posting more information about the patches that I am working on in regards to this soon.
I think that by now though, I can say that Amon Tobin got his sound with the use of Kyma and a Haken Continuum. The two go hand in hand. For those of you who don't know what these are here's an explanation:
Kyma
Basically a modular synth platform. I use the term loosely to refer to a platform that allows you to build synths and FX (in the same way as Max For Live). It seems to be about the most powerful one out (at least if you believe the hype). It's been around since the late 80s. The downside about this thing is that it relies on external hardware. The box is called pacarana. It starts at about $3000 USD.
Haken Continuum
This is a midi-like controller. It is pressure sensitive and works like a fretless bass (i.e. you can slide between notes). It is also depth sensitive so the harder you push, the more velocity or whatever. See Amon Tobin's ISAM website for his demo. A basic one of these will set you back about $5000.
So, if I wanted to outright steal Amon's sound, I would be looking at $8000 for starters. But, then, I'd have to learn how to build patches for the Kyma system which would be similar to building Max patches I guess.
If the history of computing has taught me anything though - it is that when a company claims a position as a sole arbiter/proprietor of a concept/technique, it doesn't take long before the others catch up. In saying this though, there seems to be no consolidated platform that has what Kyma offers out of the box. Max For Live can already do most of the forms of synthesis and FX that Kyma can, but it seems that the patches built for Kyma are further down the road and more of the functionality is consolidated. If you talk about the various interesting forms of synthesis that I have been talking about (granular, spectral harmonics, vocoding, etc) - these can all be done in Max For Live/Ableton. However, they tend to come in bits and pieces in the user library. I haven't yet come across devices which allow you to combine these technologies.
The question now is not a specific question of "How To Do Spectral Morphing"; it is a question of whether or not Kyma, really is so far ahead that no other platform will get close. And, could we achieve the same kinds of results with say Max For Live, or some other synth platform? My gut feeling is that just about everything that Kyma does can be done in Max or in other platforms. However, there's a big difference between "CAN" and "IS BEING". I just don't think anyone is using Max in the same way as Kyma is being used right now.
Anyway, the self-imposed task I have set for myself is to answer the two questions above. I would like to turn this thread in to an open ended discussion on these topics. So, any points of view are welcome on these questions. I will kick this off by listing the forms of synthesis that Kyma does (http://www.symbolicsound.com/cgi-bin/bi ... Algorithms). I would like to know if these can be done in Max For Live, Ableton, or some other synth platform. They are bold if I don't think they can currently be done. Italic if I don't know what they mean.
Synthesis:
additive
TAU
aggregate
cross-filter
waveshaping
modulation
distortion
formant
granular
multiwave
subtractive
FX:
Doppler-shifting (both stereo and quadraphonic)waveshaping-distortion
granular reverb,
live spectral morphing (not a simple parameter cross-fade, but a true analogy to a visual morph)
several flavors of highly intelligible vocoding
cross-synthesis and cross-filtering
freeze-framing
time-stretching
TAU analysis/synthesis
slipstick synthesis - No Idea what this is but I think Symbolic Sounds are patenting it
modal filter physical model
live spectral analysis What is this anyway?
I think that by now though, I can say that Amon Tobin got his sound with the use of Kyma and a Haken Continuum. The two go hand in hand. For those of you who don't know what these are here's an explanation:
Kyma
Basically a modular synth platform. I use the term loosely to refer to a platform that allows you to build synths and FX (in the same way as Max For Live). It seems to be about the most powerful one out (at least if you believe the hype). It's been around since the late 80s. The downside about this thing is that it relies on external hardware. The box is called pacarana. It starts at about $3000 USD.
Haken Continuum
This is a midi-like controller. It is pressure sensitive and works like a fretless bass (i.e. you can slide between notes). It is also depth sensitive so the harder you push, the more velocity or whatever. See Amon Tobin's ISAM website for his demo. A basic one of these will set you back about $5000.
So, if I wanted to outright steal Amon's sound, I would be looking at $8000 for starters. But, then, I'd have to learn how to build patches for the Kyma system which would be similar to building Max patches I guess.
If the history of computing has taught me anything though - it is that when a company claims a position as a sole arbiter/proprietor of a concept/technique, it doesn't take long before the others catch up. In saying this though, there seems to be no consolidated platform that has what Kyma offers out of the box. Max For Live can already do most of the forms of synthesis and FX that Kyma can, but it seems that the patches built for Kyma are further down the road and more of the functionality is consolidated. If you talk about the various interesting forms of synthesis that I have been talking about (granular, spectral harmonics, vocoding, etc) - these can all be done in Max For Live/Ableton. However, they tend to come in bits and pieces in the user library. I haven't yet come across devices which allow you to combine these technologies.
The question now is not a specific question of "How To Do Spectral Morphing"; it is a question of whether or not Kyma, really is so far ahead that no other platform will get close. And, could we achieve the same kinds of results with say Max For Live, or some other synth platform? My gut feeling is that just about everything that Kyma does can be done in Max or in other platforms. However, there's a big difference between "CAN" and "IS BEING". I just don't think anyone is using Max in the same way as Kyma is being used right now.
Anyway, the self-imposed task I have set for myself is to answer the two questions above. I would like to turn this thread in to an open ended discussion on these topics. So, any points of view are welcome on these questions. I will kick this off by listing the forms of synthesis that Kyma does (http://www.symbolicsound.com/cgi-bin/bi ... Algorithms). I would like to know if these can be done in Max For Live, Ableton, or some other synth platform. They are bold if I don't think they can currently be done. Italic if I don't know what they mean.
Synthesis:
additive
TAU
aggregate
cross-filter
waveshaping
modulation
distortion
formant
granular
multiwave
subtractive
FX:
Doppler-shifting (both stereo and quadraphonic)waveshaping-distortion
granular reverb,
live spectral morphing (not a simple parameter cross-fade, but a true analogy to a visual morph)
several flavors of highly intelligible vocoding
cross-synthesis and cross-filtering
freeze-framing
time-stretching
TAU analysis/synthesis
slipstick synthesis - No Idea what this is but I think Symbolic Sounds are patenting it
modal filter physical model
live spectral analysis What is this anyway?
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
OK. Just stumbled upon this dissertation. Get ready for some heavy but enlightening reading:
http://www.klingbeil.com/data/Klingbeil ... on_web.pdf
http://www.klingbeil.com/data/Klingbeil ... on_web.pdf
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
Holy @#$@ing #$@#! This guy has done some serious @#$#. Check out SPEAR!
http://www.klingbeil.com/spear/
You can actually pull apart the various part of a sound. I haven't played with this much yet, but I was able to separate two parts of a vocal sample. I bet you could easily morph samples with this. Holy @#$#!
http://www.klingbeil.com/spear/
You can actually pull apart the various part of a sound. I haven't played with this much yet, but I was able to separate two parts of a vocal sample. I bet you could easily morph samples with this. Holy @#$#!
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
I just made a MaxForLive patch which does spectral harmonic pitch shifting. This is part of the while unravelling puzzle:
http://www.maxforlive.com/library/device.php?id=895
http://www.maxforlive.com/library/device.php?id=895
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
From the blurb, that spear thing sounds like another interface on a phase vocoder. Which essentially just breaks down a sound into lots of sinusoids. And I suspect doppler shifting might be the kind of stereo fm you can achieve with e.g. waves mondomod.
Re: How To Do Spectral Morphing
I'm getting great results with the Max Patch I made today. The concept it very simple. Essentially, I created an Autotuner by accident. I wouldn't have though that something as simple as this would be useful for what I am trying to do. But, I guess what I'd really never thought about was that up until now, it was never really possible to get good harmony with a sample because if you play one sample at different pitches, the sample plays at different speeds. FFT or spectral harmonic pitch shifting clears up that problem and paves the way for doing so much more with organic sounds which is what I've been looking for.
I can't believe this is not a standard featue of DAWs or at least samplers. Have I been missing something? Are there features to do this lurking in Simpler/Sampler?
Anyway, judging by the results so far, this could be the key to understanding all my problems.
I can't believe this is not a standard featue of DAWs or at least samplers. Have I been missing something? Are there features to do this lurking in Simpler/Sampler?
Anyway, judging by the results so far, this could be the key to understanding all my problems.
Download and listen @ http://www.syncretia.com
Re: How To Do Spectral Morphing
That sounds awesome, Krudd. I'll check out the patch when I get home from work.
re: doppler, I believe Robert Henke has a Max patch that roughly approximates a doppler effect. It may or may not be what you're looking for, but I've got it and it's pretty fun.
re: doppler, I believe Robert Henke has a Max patch that roughly approximates a doppler effect. It may or may not be what you're looking for, but I've got it and it's pretty fun.