This paper presents the classical methods used for finding the induction machines parameters; then an original method conceived by the authors is detailed. The method consists in the acquisition of the voltages and currents corresponding to the phases A and B and also of the speed, during the no-load starting transient process of an induction motor. The mathematical model used in this situation is the one containing the voltages and fluxes equations, in the two- axes theory, with per unit quantities and in a reference frame solidary with the induction machine stator, to which the motion equation is added. The used objective function is built for having minimum value when the mathematical model answer is closed to the data obtained experimentally. In other words, it must be minimum when the areas delimited by the graphics of the longitudinal and transversal components of the stator current phasor (determined experimentally and by computation), respectively between the ones of the speeds (determined experimentally and by computation) are minimum. The advantages of the new method over the classical methods are emphasized with the help of a graphic comparison. The method proposed for the determination of the induction machine parameters, has the following advantages: it provides an estimation precision which is superior over the similar methods tackled in the literature; it may be successfully used both for the determination of the squirrel cage induction machine parameters and of the phase-wound rotor ones; it requires a very low energetic consumption.