At first sight, the GUI is a bit puzzling. The lack of contrast and the tiny inscriptions might be problematic if you work at a reasonable distance from your monitor. Fortunately, a pop-up with a green background indicates the name and the value of a parameter when mousing over it. It would be better if it was bigger...
However, the different sections/modules are placed according to the signal path. There are two menus, one for global help when hovering the mouse cursor over the plug (activation by the “question mark” sign), the other gives information on the speakers, both of which are very helpful. Another good point: you can use the mouse wheel to adjust values or scroll in (almost) all of the plug-in areas.

Speakerphone is broken down into three parts: speakers simulations, effects and samples. In the upper section you’ll find the presets bar, giving access to up to 500 carefully tailored settings. On the left there is the Gain knob and its Auto Level, which ensures that the input level is always optimized. On the right are the Out level and its limiter.

Then you’ll find the speaker section which has, like each module of the plug, its own bypass. Hence you may choose to use only reverb or effects, or activate a module when required. Any Speakerphone function and parameter can be attached to a MIDI controller by right-clicking (or ctrl-clicking) on it, which opens a MIDI Learn/Reset menu.

Six icons are shortcuts to the speakers classification and a central monitor shows a picture of the selected speaker. You can choose between mono and stereo, pre- and post-FX. Just below there’s two Show/Hide control tabs.

Audio Ease has implemented a nice function, Kick. Below the Dry/Wet slider, there’s another one being flanked by two arrows: when you click on these arrows, the slider will move from Dry to Wet (or vice-versa) according to a period of time between 0.10 and 30 seconds (by intervals of 0.02 seconds when pressing Alt). It’s not only a static balancing between Dry and Wet signals, but kind of dynamic modification of each module independently. This is a clever function, allowing the user to perfectly simulate changes like someone walking from one room to another, or two people talking on either end of a phone. The required precision can be achieved by automating the function.