In this paper we deal with switching fuzzy filtering design problems for a class of discrete-time nonlinear stochastic systems with time-varying distributed delays. To investigate the system stability, the considered nonlinear stochastic systems is presented by switching fuzzy stochastic models. To relax the stability conditions, an improved free-weighting matrix technique and the piecewise Lyapunov- Krasovskii function are employed. Some new delay-dependent sufficient conditions are established to ensure that the filtering error system is stochastically stable with a guaranteed H∞ performance index. By using a novel matrix decoupling technique, all sufficient conditions can be expressed in terms of linear matrix inequalities (LMIs) and therefore are numerically tractable, as shown in the example included in the paper.