Based on a represention of F0 in mora units (F0mora), several parameters and methods were introduced in order to identify accent types of accentual phrases: new candidates for F0mora estimation method; the addition of relative F0mora parameters and relative power parameters; and a new accent type identification method. The candidates for F0mora were investigated in order to find the best matching with the perceived pitch values. As for the relative F0mora parameters, new delta-F0mora parameters were proposed, in order to take phrase contextual effects into account, and to supply additional information in segments with missed F0 data (like in devoiced vowels). Relative power parameters were also investigated, because power also seems to influence in the accent type identification. As for the identification method, neural network models were proposed to find a suitable weighting for each parameter, and a transformation of the input parameters were proposed using Gaussian distributions in order to deal with the parameters with missed data.