Hi all - first post, hope you don’t mind a slightly long-winded question. I’m looking at Wavelab for a very specific purpose that not much else seems suited for at this point. My aim is to split a longer recording into shorter .wav files for further processing and importing into Kontakt. In order to use Kontakt‘s auto-mapping function, I need to embed the root key or MIDI number in the file name (I believe it also reads metadata, but the filename method seems tidier). I understand that Wavelab can detect pitch and do all this, but to what extent can this be automated? Here’s my proposed workflow:
Record a single pass on an instrument in a DAW, e.g. every third note on a violin at a given dynamic.
Export as one audio file and load into Wavelab
Automatically - so far as possible - slice each note into its own clip/region, remove silence, detect pitch, and export individual .wavs with that embedded in the file name.
I’ll grab the demo and try this out at the weekend, but it would be great if someone could let me know if I’m likely to be barking up the wrong tree already! It’s a little difficult to get a sense of exactly how all these components fit together.
Bonus wish: If this is indeed possible, can it be done A. from a watch folder, and B. on more than one file at once (e.g. in the case of multiple velocity layers)?
Okay, I found some videos showing almost exactly this functionality. Perfect. Still would love to hear any perspectives on the “automatability” of the above, plus any other insights from people using it in a similar way. Optimistically dreaming of a watch folder named “do all the tedious sampling editing stuff for me.”
Interesting topic. I was doing something similar couple of days ago. I wanted to sample an instrument with several microphones (3) so I decided to record single multichannel audio file directly to Wavelab. The idea was to edit and cut the samples at once and split to individual mono files at the end.
The first problem was the autosplit function doesn’t work with multichannel files, only mono or stereo. So I created markers where I wanted to split the samples and rendered as multi mono. I named the markers as they would be used later for file names when splitting the samples.
Rendering to multi mono, I found that markers are only copied to the first of the resulting mono files. I had to copy the markers manually. Fortunately it’s relatively easy to do, just copy all markers and paste them to the other files.
Now it’s time to split the samples. This bit was easy as I already had markers in place. I used the autosplit according to markers. I find this is the safest options. This is where I tried to detect the pitch of the resulting samples. Well, it was correct most of the time but not 100%. So if you rely on this then better check it afterwards.
Then I normalised all the samples. This was done using batch processor once all the samples were already cut.
So this is my experience with sampling in Wavelab. Maybe not what you hoped for in terms of everything being automated… But to be honest I didn’t fully explore all possible options.
It was a lot of videos each showing little glimpses of what I want really. This one about auto split is what convinced me this fundamental “split files into correctly named samples” workflow ought to be possible: Auto Split in the Audio Editor Wavelab 10 Pro - YouTube
Interesting thoughts @misohoza, thanks. Even an imperfect starting point could save a lot of time. I think most of my current project will be done with two mic positions on each instrument, so stereo files should be workable for now. Recording directly into Wavelab is something I hadn’t considered, but it sounds ideal for quick and dirty Kontakt instruments. Also more fully-featured ones, I assume, but it’d take me a while to get up to speed.
Thanks for posting the video. Nicely explained. I would probably run the autosplit twice. First do a test run and choose the create markers option. See what you get and possibly fine tune the settings or nudge some of the markers if needed. Then run it again this time cutting at markers you just created.
You can also assign the key to samples by setting the first one and then choosing the increment. This is handy if you sampled chromatically or 2, 3, 4… samples per octave. So you don’t need to depend on the automatic pitch detection.