On a live recording, there’s a resonance in the room, PA, etc that sounds like a peak in the low mids when the singer hits that tone. Using Audacity 2.0.3, I’d like to EQ that peak out.
I’m thinking I’d need to watch a spectrum display at the particular point in the recording where the “woof” occurs, figure out the center frequency and width of the peak, and set a multi-band EQ accordingly. Is anything like that possible, or maybe not too difficult?
Usually it’s best to simply experiment. Pull down one EQ slider (in the Graphic EQ Mode) at a time (or a few at a time) to zero-in on the problem frequency. Pros sometimes sweep a notch filter (or sometimes a bandpass filter) to find the problem frequency (or frequencies). But since Audacity doesn’t do real-time effects, that’s not too easy in Audacity. But it’s not too hard to try all of the EQ sliders in the low, mid, or high range wherever you suspect the problem lies.
Room acoustics can be tough to filter-out. Pro live recordings are generally close-miced with separate mics to pick-up applause an room ambience. Some acoustic live recordings are near-miced with mics at about the conductor location. It’s rare to place the mics in the normal audience listening position because the amount of room sound that sounds great live doesn’t come across very well in a recording.
There’s an old trick of crumpling a newspaper in front of the microphone. That’s not a totally dreadful “white noise” generator. You can also listen to the mixer while you whistle or hum or woooooof. You get to one pitch tone and it’s suddenly painful.
On a live recording
Meaning you only have the recording, right? No room? You should be able to listen to the performance and mark the time when a particularly evil honk occurs. Drag-select around that portion and Analyze > Spectrum or change the timeline view to Spectrogram.
You can help with the equalizer, but resonant peaks also ring. As a bell. So some of them have a boost and a tail. That’s harder to get rid of without damaging anything else.
Thank you both for the advice. I’ll try these methods out this weekend and let you know.
This was a recording of dinner Jazz with Piano, a delightful singer, and myself on Upright Bass. My Zoom H2n recorder on the closest table to us, and a Bose Compact for sound. Lots of audience noise of course, but I’ve learned to get the Zoom as close to us as I can.
The recording’s ok for what I intended, but if I can reduce that resonance a just bit, it’s worth a try.
If you just use equaliser to attenuate those resonant frequencies they will be attenuated across the entire duration of the track , whereas the compressor only attenuates when the volume of those frequencies goes above a user-defined threshold.
This is the model that’s still going for new prices, even though it hasn’t been made for a while.
I’ve been following him for a long time. He took the unprecedented step of having the same name as a popular country and western singer, so he’s impossible to Google. Then, recognizing that, he picked a professional name impossible to remember.
I don’t remember what it is.
In all that time I don’t remember him ever trying to personally capture a performance in front of an audience. You may be experiencing why. He does have a history of never posting a crappy performance, although I’m sure there have been some. So to me he appears to be a recording deity but I’m sure he’s sitting on a pile of failed recordings.
Best place for them.
Nobody asked you to post a sample of the work. Sometimes we can just tell you what to do. You can only get through about five minutes of stereo WAV on the forum, but if you have a DropBox account or equivalent, that’s easier.