As long as there is no phase transition, the DSC signal is equivalent to the difference of the heat capacities of the sample and the reference – theoretically! Practically the finite heat conductivity may cause deviations. It is therefore advisable to calibrate with samples of known heat capacity.

You integrate the heatflow on the time and draw the data in function of the sample temperature, if you have a phase-change you will get a step, if not only a more or less linear curve (maybe even a plateau). The height of the step represents the latent energy, the slope of the curve is the sensible heat capacity. You divide by the mass of the sample to have the specific values.