VSTi -> sf2

Audio Plugin Hosts and other audio software applications discussion
Post Reply New Topic
RELATED
PRODUCTS

Post

Hello

New version of VST dumper was released. Now it can create sf2 files from VSTi preset.

http://audiolabz.com/vd02.rar

Greetz

Post

Just tryed it out, very good and can be very useful.
Thank you for this post! :shock: 8) :D
Image

Post

This sounds like a great idea.
Good stuff!

Caleb
Happiness is the hidden behind the obvious.

Post

idea -> good
wav export -> hmm... no wavs exported
sf2 export -> ok, but
Why multiwave sf2 create other wavs by resampling first sample???
That's the same, as i export only one wave and play with it in any sampler... except of file weight...
All waves should be rendered separately and with assigned root keys.
Custom map or interval switch (between notes to sample), may be nice feature.
File names inside sf2 need rename imo or maybe simple "space" between root and stereo channel.
There is no way to open gui - often thats the way to load presets.
U should let it import fxp/fxb.
Now it can use only patches included with plugin by default.

Anyway good work
Cheers

Post

wav export - i missed one 'if', now wav export works (http://audiolabz.com/vd02c.rar)
sf2 export - already samples are rendered separately for notes specified in note range (you must uncheck 'single note' option) - (updated in 0.2c version, also sample clipping, volume ramping, conversion to 16bit with dithering, and about 1s break between next 'note on' added (in next versions that feature will be editable))

Options like vsti editor, fxb/fxp loader and note range editor are under developing :) Be patient.

Greetz

Post

Not a bad idea, but:
Xlutops chainer (even the demo) does everything way better.
- You can define the notes that are sampled.
- "Duration" parameter is working. With VSTi dump it doesn't seem to work properly.
- You can load presets in Chainer. With VSTi dump the default ones are used, from what I can see.
- You can try out the patch in Chainer.
- Notenames in Chainer are correct. In VSTi dump, exported waves without a sharp or flat have an additional "-" which makes them completely unsueable for any automapping sampler. Filenames should have some "C0" add on, rather than "C-0".
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

Hmm i tried to load sf2 exported with chainer into Vienna SF studio and i can't :)
Anyway i try to do something similar to chainer

Greetz

Post

Ok, Mate.
In vd02c new version all's ok. Good work.
About resampling... in previous version all keys root was C5...
nevermind - now its ok.

cheers.

Post

Hello

Now i am integrating vsti dumper to my .LOVE framework, and adding new
features:
-note selector
-multisample or chord mode
-audioLabz directSound & ASIO host for preset listening and live playing

Probably i will add stuff like:
-multiple vsti dumping -> for creating new sounds Smile
maybe with vst fx chaining
-envelope editor
-some filters

Screenshot of initial 0.3 version: http://audiolabz.com/dumper.jpg

Greetz!

Post

That's better.
You should integrate note velocity as well, makes quite of a difference for some patches.

Anyways, may I ask you why you do it?
I mean, yes, it's a nice project, still, Chainer really does everything (even the demo).
If you want something the Chainer demo can't do, add saveability of patches to it, once you got the possible layering done.
Oh, and VST effect support wouldn't be bad either.

Sorry to be such a prick, but really, right now Chainer does everything you've implemented so far - plus a bit more.
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

Sascha Franck wrote: Sorry to be such a prick, but really, right now Chainer does everything you've implemented so far - plus a bit more.
if this were applied to every audio tool, would we still have 500 compressors, eqs, sequencers, etc, etc, etc ?

quantity or quality

having more is almost always good or is it ? :)

Post

Sascha is right.

How about AUTO-CROSSFADE looping ?
ADSR rebuild from the VSTi patch ?
(at least activate articulation,
so ADSR will work right away with sfz.)
AUTO-VELOCITY layers ?
Opening the VSTi gui of the loaded VSTi ?
Set note sampling to 2,3,5,7... steps ?
(so every 3rd note ect. gets sampled.)

Then it is really getting interesting. :wink:
Last edited by Midiworks on Wed Nov 24, 2004 2:52 pm, edited 1 time in total.
Image
DSP with attitude

Post

nool wrote: having more is almost always good or is it ? :)
In this case, at least so far, no.
Compressors all sound diferent, but VSTi dump doesn't do anything special.
As said, I'm being a prick here, but well...

As far as Midiworks' suggestions go: An automatic loop point insertion process would be killer for anything pad-style. One would perhaps enter the start and end points and the app would automatically insert some x-fades and loop points.
Still, it might not work on a lot of modulating patches, but it might be worth a shot.

Oh and yes, auto velocity layers would be cool too. Not THAT much work to export at different velocities in chainer, but in case you'd be extracting a LOT of velocity layers it would be nice not having to export them all separately.
Something meaningful like "V90", "V120" should then be added to the filename.

Talking about filenames, I allready mentioned it, note names should be represented in the form "WaveXYZ_C1.wav", "WaveXYZ_F#3.wav" etc. No further "-" and so on (MIDI note name conventions don't contain those, so automapping, if possible in your sampler of choice, won't work).
In addition (or alternatively), the root info could be embedded into the file header.
There are 3 kinds of people:
Those who can do maths and those who can't.

Post

Sascha Franck wrote:An automatic loop point insertion process would be killer for anything pad-style. One would perhaps enter the start and end points and the app would automatically insert some x-fades and loop points.
Still, it might not work on a lot of modulating patches, but it might be worth a shot.
It can work very well,
Disco DSP's Highlife proves it.
Simply chose a fixed crossfade lenght for a
pre-specified sampling duration.
You can look at the times in the Highlife demo. :wink:
Image
DSP with attitude

Post

yeah - coll stuff :D ...absolutely useful...thx

tro

Post Reply

Return to “Hosts & Applications (Sequencers, DAWs, Audio Editors, etc.)”