I have a very old copy of my grandmother singing with piano accompaniment. The original music came on a cassette tape which I imported to my computer. I would like to extract the voice only and ditch the piano. Then I am going to mix her voice with some orchestration. What is the easiest way to accomplish this? If some piano remains, that is okay as well. I just want mainly the vocals.
It’s not that easy. If you had a super high quality, stable stereo performance and a bunch of other moons and stars lined up, you may have a fighting chance. but starting with a cassette tape…
You can’t use straight sound filters because the piano and voice have similar overtones, so you’re left with the stereo imaging tools…
One of these I believe will do both isolate and remove if it’s possible with your performance. Both of these tools depend on a stereo performance. If you have a mono performance–even if it’s two-track mono, you’re dead. If you listen carefully on headphones, is the piano coming from one side and the singer from the other, or are they both coming from straight ahead?
First task though is to capture the tape into Audacity and export that as a master copy WAV (unless you already have it as a WAV) file AND make a separate backup copy on a separate disk (e.g.extenal USB/firewire disk say).
Do NOT edit this master copy - rather take a copy and do your editing work on that.
That way you can always return to the original saved master and start over with the editing - this is precious un-repeatable material.