Since just about every artist, songwriter, musician and producer now has a home studio, most everyone has run into the problem of finally getting a perfect take only to have it ruined by some outside noise. Sometimes you get so desensitized to the noise that you don’t even hear it until hours or even days later when it jumps out like a sore thumb during playback. There are a lot of tools available to minimize or even eliminate background noise, but most of the good ones are either complicated or expensive (often both). That all might be in the past though, as Waves has just introduced its new AI-driven Clarity Vx noise removal tool that’s truly a game changer in cost and ease-of-use.
It’s All In The Neural Network
Clarity Vx is aimed primarily at voice, vocals and dialog that have been seemingly ruined by hiss, hum, or even harsh external noises. From light air conditioner or computer fan noise, to heavy traffic outside your bedroom studio window, the plugin is a whiz at either eliminating it, or at least decreasing it to a level where you can’t hear it in the track.
The plugin is powered by Waves Neural Networks® technology that’s been fed with hundreds of thousands of audio files of vocal recordings, and trained through the combination of machine learning and human evaluation to recognize the difference between voice and ambience. That means that, unlike other AI-based approaches, Clarity Vx doesn’t have to spend time learning your track first before it can do its thing. It works instantly.
Simple To Use
When I mentioned that it was easy to use before, that was no exaggeration. At its most basic, all you have to do is select from one of the two neural networks and turn the big control in the center of the plugin until the noise decreases. In most cases, the noise goes away like magic. A round display to the right shows how the signal is being analyzed before processing.
The two neural networks have different purposes. Broad 1 is better at preserving both the main and secondary voices, when more than one voice is present in the recording. Broad 2 is recommended for severe noises. It’s also better at separating the main voice from secondary background voices, when more than one voice is present in the recording.
There’s also a Reset button to clear the neural network history cache. Certain events, such as an abrupt, substantial change in the noise profile, a quick change of speaker, or a sudden loud noise, can cause the network to lose focus or compromise quality, and Reset overcomes those problems without losing your settings.
More Control With Stereo
A stereo signal brings up a couple of additional parameter controls. Analysis controls how a stereo signal is analyzed before neural network processing. It has two selections: Single sums the left and right input channels before analysis so both channels are processed using the same analysis data.
Double is used when the left and right channels are significantly different as it analyses the left and right channels independently and processes them separately. This results in more precise processing for each channel, but it requires substantially more CPU. There’s also a Width control to further adjust the stereo image.
Although Clarity Vx is meant to be used primarily with voice, it does work on other mix elements as well. For those tracking in the studio with a full band, you’ll be happy to know that the plugin does a pretty good job of decreasing or eliminating leakage.
Clarity Vx is not only simple to use, but you can’t beat the price either. At just $29 with the code CREATE40, it’s a no brainer addition for your plugin collection. There’s also a pro version for heavier duty use that’s $249, but the Vx version should be sufficient for most music uses. Be aware that it does take up a lot of processing power and has a lot of latency, so be sure that delay compensation is engaged when you insert this plugin.
You can find out more here, or watch the video below.