Not sure exactly what you want to accomplish, if you're talking about the actual amplitudes, you could try processing the audio you're feeding it as a source, maybe adding a reverb, or that sort of thing.
If you're talking about the shape of the output, try the various matte choker effects to see if they could help.
A picture is worth a thousand words.
This is not just Audio Spectrum. It may not have been used at all. When you see an effect that you want to duplicate you have to start by breaking down the shot into the elements that you see in the frame. Here I see a background layer that looks like it was mirrored. The background is static. Above the background is a snow effect that could have been applied to the background layer but would probably be easier to handle on it's own layer. The snow also appears to be mirrored. The snow expands from the center so that is probably a particle system. Then there is a layer that is made up of posterized colors that react to audio suggesting that maybe Audio Spectrum may have been used as part of the effect. Then there is a circular logo on top. The logo and the posterized colors both scale in sync with the audio. Let's start breaking down the shot.
Background - Split the image in the middle and mirror vertically.
This can be done by masking and duplicating or by using Effect>Distort>Mirror
Snow layer - Just about any particle system or snow effect + mirror
Posterized color layer - We already know that Mirror is the last effect in the stack. Audio spectrum can be used to create a smoother shape by increasing the number of frequency bands and the frequencies can be adjusted to give you a smoother look. You can mess with the controls until you get kind of the shape you want and then adjust the inside and outside color to give you kind of a gradient. You can then apply Colorama or Posterize colors to get the look that you want. Without some serious EQ to an audio track to set up your reference audio I'm not sure you are going to get what you want. Another approach would be to use Sound Keys from Trapcode or the Audio to Keyframes function from the Keyframe assistant to get a set of keyframes that you could use to control the position properties of the start and end values of a gradient fill on a shape layer and then use something like turbulent displace and expressions to distort the gradient shape layer.
The last part of the puzzle is the scaling synced to the audio. You would use Audio to Keyframes here and an expression to control the scale of the logo and the posterized colors layer. The expression would look something like this:
t = thisComp.layer("Audio Amplitude").effect("Both Channels")("Slider");
s = linear(t, 0, 45, 80, 100);
I threw something together very quickly using Audio Spectrum. It's not exactly the same look but it shows the concept. I think that the second option that I talked about could give you a better look to the posterized color layer. I don't have time to fuss with that right now.
I think you'll get closer to that look by starting with a greyscale gradient filled rectangle + Colorama + distortion effects like Bezier Warp that can be controlled by tying position or offset values to the Audio Amplitude values with an expression in the same way you can tie scale to the audio above. When you get a nice rainbow that is distorted by the audio you can use Polar Coordinates from the Distort section to convert that to a circle and mirror to reflect it. That might be fun exercise for me when I get some more time.