Vocal separation produces the complete vocal track and an instrumental track. Acapella extraction keeps the vocal output; removing vocals keeps the instrumental. These are different ways to use the same two-track separation.
Six-stem separation produces vocals, drums, bass, piano, guitar and other instruments. Choose it for remixing or learning an instrument; a two-track vocal split cannot give you isolated drums or guitar.
To keep backing singers in a karaoke track, remove the lead vocal with the dedicated karaoke model. To remove all singing, use vocal separation. Lead/back splitting first isolates all vocals, then separates the lead from the backing voices.
Noise reduction and reverb removal solve different problems. Preview one operation at a time and compare it with the source. Stronger processing can remove detail, and breath or wind reduction is not guaranteed for every recording.
AIstemify uses publicly released pretrained separation models, including BS RoFormer and the Anvuew karaoke model. We do not claim to have trained these models. Different tools select different outputs or processing steps.