The frequency response curve that you got after you measured your room is what our software has picked up to be the average “character” of your set-up. It shows peaks and valleys across the frequency spectrum from 20Hz to 20kHz (typical frequency range with sample rate at 44.1kHz).
Any of the peaks and dips in the curve shows where the audio you hear is being masked by your room. Dip around 100Hz will seem like your bass has no “body” whereas a peak at 7kHz and higher will make you process your cymbals/vocals to be darker than they need to be.
Even though it’s tough to point the exact threshold at which the dip/peak is considered a significant one, you should look at the graph as a unified thing and any outlying differences from the “natural” flow of the graph will poke out as a splinter in your finger.
After you save this calibration file you just created, it’ll take the graph you saw and flip it, essentially neutralizing every dip and peak that it picked up, giving you the natural sound of your audio.