Drying kinetics of carrot cubes were carried out at 1 m/s air velocity at different air drying temperatures (30, 40, 50, 60 and 70±0.1 °C) (AIR experiments), and also at the same experimental conditions but applying high power ultrasound (US experiments). Two kind of diffusion models were used to simulate the drying kinetics, according to external resistance to mass transfer being considered (ER model) or neglected (NER model) for solving the diffusion equation. Diffusion ER model was solved using a finite difference method. Drying rate increased as air temperature was higher. Ultrasound also increased drying rate at the different temperatures, but the improvement on drying rate decreased at high temperatures, and almost disappeared at 70 °C. Effective moisture diffusivities only showed an Arrhenius type relationship with temperature for AIR experiments. The NER diffusion model was not accurate to simulate the drying kinetics at any experimental conditions tested. However, diffusion ER model provided a high closeness between experimental and calculated drying data (VAR>99.80). Through the parameters identified of the ER diffusion model, effective moisture diffusivity and mass transfer coefficient, the influence of the power ultrasound application on internal and external resistance to mass transfer was shown to be significant (p<0.05).