I don’t think anybody can do that. Audacity has sound activated recording, but it responds to anybody speaking – or any sound, not just one person. Audacity also doesn’t do many tasks in real time, so you can’t apply filters or effects.
This is a fairly common question that involves thinking. Software, as a rule, doesn’t think. This is what kills people trying to filter out a TV set from a live voice recording. I know you can tell the difference because you can think about the announcer on the TV that somebody left running by accident, but all software knows is all the sound in the room needs to be recorded – worse, when it needs to try to split apart two live voices. Extraordinarily difficult even for top drawer applications.
As an extension of that problem, filter out my voice only when I’m describing my car, or only when I use metaphor. You can tell the difference, but most common software is lost.
You can do it mechanically. Capture the performance and cut it word by word, but you can’t easily separate the voice from the music if they persent at the same time. Koz
You can try the vocal isolation software, but all that does is split sounds in the middle of the stereo field. That means the lead singer will leak through.
Thanks for trying to help. It was not right when I said I only wanted the womens voice recorded I also want the mens voice she is interviewing. But don’t need the first mans voice because it’s radio news and ALL the music should also not be recorded.
But wouldn’t it be possible to click start and pause whenever music start and at the end export the recorded file and put it online on Internet for people to hear? But it’s not easy and that’s why I want to pay someone to do it for me.