I have a recording of a voice conversation (panel) made on a personal voice recorder (Sony IC recorder) which unfortunately was positioned close to the music speaker - something we forgot to test. Much of the conversation is obscured by music (beats, for example) and there are some portions we completely don’t understand and cant’ transcribe. I recorded portions of this audio into audacity and then tried experimenting, mostly with the High Pass Filter. It helps, but not much. What other techniques can I use to dim as much of the music as possible to focus on spoken voice behind? Thank you very much in advance.
You might benefit from the Telephone Filter. In Audacity 1.3, run Effect > Equalization… and drop everything below 300 and above 3000, so the graph looks like a haystack with a flat top. Accuracy isn’t important. Apply it.
That’s approximately the classic telco frequency response so as to make human speech as intelligible as possible on the phone.
That’s pretty much the ball game. If that doesn’t help, you have no show.