If you have a detector (for example, a diode array detector), you can play games with the spectrum to get that data. If you are going down the spectrum route, it is better if you could find an spectral absorbance in one compound that isn't in the other. Say if one compound absorbed at 325 nm and the other absorbed below 254 nm. Then the calibration can be relatively straightforward.
Another way many to get this data is using spectrum subtraction. You subtract a reference spectrum from the data spectrum. The right software can do it. It seems to work best when the two compounds are near the same intensity. Since this is rarely the case, I was never too impressed with the results.
A better approach is to change your chromatographic conditions so the two compounds don't elute at the same time (i.e. a different column, different solvents, or a different buffer). That can be a lot of work.