Direkt zum Inhalt

I just imported a complex midi file from Melodyne DNA and things are a bit messy now

Di., 15.03.2016 - 12:53 Permalink

It worked with solo instrunents before v2.6 , but there may be progressive improvement in the recognition and removal algorithms in subsequent versions, 

As I said earlier, I haven't used it myself, because it's currently Mac only (but Win version is underway). I've seen it in action - very impressive. However, it does require substantial effort.

Do., 24.03.2016 - 23:14 Permalink

It worked with solo instrunents before v2.6 , but there may be progressive improvement in the recognition and removal algorithms in subsequent versions, 

As I said earlier, I haven't used it myself, because it's currently Mac only (but Win version is underway). I've seen it in action - very impressive. However, it does require substantial effort.

I just tried ADX TRAX PRO now and there is not much difference to Spectra Layers Pro besides minor tools when  the main task is not extracting vocals which ADX does by itsself. The extraction of vocals works very well automatically in ADX but as soon as manual instrument selection starts I have to draw the pitch by hand which is as hard as drawing into the spectrograph in  Spectra Layers even so you can change the view to 3D. As soon as it gets more complex and crowded the spectral drawing provides not enough details to really draw any instruments. 

My final solution to this is to use the edit function in melodyne itself after the material was converted to the blobs used by melodyne as these are the only ones you can really use to see different instruments and melodic elements in the bass section. I am going to use the same workflow as I would in ADX TRAX PRO or Spectra Layers and delete / move everything out of the way that is not the melody playing. Its much easier than selecting some diffuse spots in the spectrum.

Still it was very interesting to try these tools and maybe some of you have hints about me overseeing something obvious in ADX TRAX PRO or Spectra Layers. I also still wonder if the ADX tool used to autodetect the vocals can be abused to automatically select a certain instrument in the whole track which would change the workflow dramatically.