resynthesis work arounds?

[Dewdman42]

Since Arturia V does not have the resynthesis capabilities, what work arounds are people using to accomplish resynthesis, particularly the analysis of samples to create timeslice frames based on carrier+modulator?

There must be some way to analyze a sample at a given point in time and pull out the carrier harmonics needed to create that part of the sample additively.  Well, what would be the easiest tool to use for doing this outside of the SynclavierV software? 

Secondly, I'm presuming most people would likely have to do simple resynthesis using just the carrier and some fourier analysis.  Is it wishing too much to be able to analyze a sample and extract carrier+modulator?  I guess so but just asking.  Resynthesizing with just Additive timeslices is one thing, but resynthesizing using at least some level of FM would be much more difficult to do in my mind, I am just wondering if there is any way to do it some what analytically with software...  Using FM would open up interesting performance possibilities to tweak the modulator over time or with a controller, etc

SynclavierV is an amazing instrument I am blown away by it, even exactly as it is now, but I'm also extremely intrigued to do some resynthesis based on some sampled material, but I feel like I'm just taking wild guesses by drawing around on the carrier and modulator harmonics graphs.

[LBH]

I think there are applications that can extract harmonics from a sound file. And perhaps some can vrite the harmonics and their levels in numbers. But even then you perhaps can't use it 100% because Synclavier V unfortunately don't show any values for the harmonic settings. I would like values on Synclaviers harmonic settings.

If you wan't to find out you perhaps can do a internet search like this and similar: Extract harmonics from audio

Was this what you had in mind?

There are feature request about this area in the forum.

[Dewdman42]

oh yea I read all the posts on this subject from last year and I'm sure there are feature requests, but I'm looking for specific work arounds that some users might be using to explore this area.   I have several plugins that are able to do this analysis such as Harmor and Alchemy but as far as I know they don't expose that information to repurpose it in synclavier.

Is there anything out there that can do real time fourier analysis?  I would think not, but just asking..  If so then it would be possible to analysis a source sample to get a harmonic graph, then tweak the siders in synclavier while watching the analysis until it matches in the analyzer.  Just thinking out loud here.

[creart]

At the moment I'm trying to program a tool that can help in doing that... Something that would come up with a set of timeslice settings that you'd have to fill in in your synclavier V to get the sound....
It will definitely take me some months to finish but as soon as I have something reasonaby useable I will upload movies of how it works and will probably sell the program for Windows and Mac...
I will also contact Arturia to see if theres a way to input dat straight into Synclavier V - that would be awesome!

cheers Hans

[creart]

btw I'm also working on a tool to convert DX7 sounds to Synclavier...

[Dewdman42]

something like this:   https://www.youtube.com/watch?v=eAcwYr_nJuI&t=8s

[creart]

Well - I'm hoping to get my software to show you the harmonic settings graphically just like in Synclavier V and with changing sound over time, it would show settings for different timeslices with the time for that particular timeslice...

[creart]

I had never seen that Lil'Devil btw :-)

[Dewdman42]

I like what you're doing.  Lil Devil is sold by the guy that wrote Welsh's Synthesizer Cookbook and its more of a teaching device then anything.  He gives it away for free with his second book.  Only runs on windows. 

For me it just gave me some ideas about how to feed a sample and synclavier both into a spectrum analyzer that has a freeze control.  Then fiddle the partials on synclavier while hitting a midi note until the graphs kinda match.  I like his commentary about how to add in some FM modulation to get the movement in the sound and see how it kind of matches up with the graph too. 

If you're gonna write your own software for this, it will be able to adjust the partials automatically... I think that's the point.