I have a stereo track with complex music on it. it starts with an count-in metronom click on 1-2-3-4-1. At the last “1” there is a bell and already the first instruments are playing. (After this “1” the metronom was already mutet during mastering)
Now I want to remove the metronom completely. The clicks on 1-2-3-4 are no problem because the metronom is alone. But at the last “1” the music is already playing and I want to remove the bell too here. I first tried to remove it with an Notch filter with Q=5, but the number of frequencies of a bell are many. It works but I still can hear it.
Now my question: Is there a filter where i can “substract” a track from another track? I have a clear recording of the exact same bell and I know the exact position of this sound in the stereo audio.
Yes it is an (sampled) electronic bell from a Yamaha keyboard’s metronom. The sound is avaiable as a single wav-file.
See attached ogg-file:
only 200msec long
its a combination of bell and click, but i want to remove both.
Do you know an AUDACITY-Effect which already can substract two tracks?
Don’t know whether this would work…
But theoretically this could work on base of an FFT-Analysis-Algo.
There isn’t a direct way to do it, but it may be possible with a bit of fiddling. Could you also post the start of the track that you want to remove the sound from (in WAV format - use “File menu > Export Selected Audio…”)
Hi Steve,
thanks you for offering me assistence, but… I’m not really interested in sombody doing this job, but in learning how to do it by my own. The job with the bell is only a first sample for more work like this.
There is only a tight, single flash frame of “2” and there is no “1”.
5, 4, 3, [beat] [beat] Music.
Build that into your metronome countdown.
Instead of using metronome, Audacity has Generate > Rhythm Track. That gives you beat patterns.
Instead of plain tick tick tick forever, it gives you TICK took took took TICK took took took. That’s where you cut to silence. The music comes in where the next TICK would be.
If the sound in the mix is digitally-identical to the separate WAV file it can be perfectly subtracted-out.
Mixing is done by summation, and if you have two tracks open they will mix when you export. If you run the Invert effect on the bell track you’ll be “adding a negative” (subtracting).
It has to be time-aligned and level matched. In a real world mix, the levels have usually been changed, the timing might be different, and other effects may have been applied. If everything is not perfect you can end-up with two bell sounds mixed together or some other undesirable side-effects.
