I’ve recently installed a new, very powerful noise reduction software package, RX5 Advanced. It’s a 2-track spectral repair and mastering program that has some abilities that are pretty amazing. It’s not exactly a miracle worker… it can’t take poorly-recorded audio and magically make it good. But it can improve a lot of problem files, some of them dramatically.
Since I’ve just started using it, there are a lot of features that I’ve only started to explore. But what I’ve seen so far is is really fun. The display is a spectrograph, where frequency is displayed on the Y axis (up and down), time on the X axis (left to right), and intensity as brightness… louder sounds are brighter. A standard waveform is overlaid in blue, and there’s a slide that lets you adjust the display for pure spectrogram, pure waveform, or anything in between. The end result is a lot of information about your file that you can quickly and easily understand. For example, see the horizontal bands in the lower right corner? That’s a train horn in the distance. The darker bands before and after the loudest parts of the file are where I’ve applied the denoise module… this file was recorded outdoors, and there was interstate noise in the distance that sounds a lot like white noise. The program was able to learn the background noise and reduce it significantly, while not affecting the direct voice signal much at all… extremely useful.
The program has tons of other features that I haven’t had a chance to try yet but promise to be useful. For example, there’s an ambience match that can add “room tone” to a voice track that was recorded on location. I can think of a half dozen times when the de-reverb module would have come in handy.
One of my favorite modules so far is the Adaptive Phase Rotation feature. This fixes a typically pesky problem of asymmetric waveforms, where there is more amplitude on one side of the infinity line at the center of the screen than the other. Sonically it isn’t a problem and sounds fine, but it robs you of headroom… when you normalize, one side of the waveform limits the amplitude, and your signal is overall slightly lower. (Besides, it bugs me to have a lopsided waveform. I like my waveforms neat and tidy.) With this module, one click corrects the problem without affecting the sound of your signal.
Right now I’m running it on my little 2.6gHz Mac Mini with 8gb of memory. It does OK on the internal drive, but some of the processes take a few minutes on long files (like declicking a 2-sided album). I’ve ordered a 3TB 7200rpm Thunderbolt hard drive that will be dedicated to audio only, and I’ll probably up the memory to 16gb (max for the Mac Mini) first chance I get.
There is so much more that this program can do, it would take weeks to go over it all. And I’m just getting started. But it’s a GREAT tool to have in my audio arsenal, and I’m looking forward to offering it to clients (and generating some additional revenue with it.)