Speech analysis for high quality speech synthesis or high accuracy speech recognition requires realistic models not only for the vocal tract but also for the voice source. This paper presents a comparison between two analysis methods for the calculation of the voice source and vocal tract parameters for voiced sounds. The first method computes once the vocal tract parameters and then ajusts a glottal parametric model on the residual by a quadratic optimization procedure. The second method iteratively computes the parameters of the vocal tract and glottal models. It was shown that first method, though a least complexity leads to signal-noise ratio slightly lower than the second method.