This paper presents the nonlinear system identification of model parameters for a capacitive dual-backplate MEMS microphone. System parameters of the microphone are developed by lumped element modeling (LEM) and a governing nonlinear equation is thereafter obtained with coupled mechanical and electrostatic nonlinearities. The approximate solution for a general damped second order system with both quadratic and cubic nonlinearities and a non-zero external step loading is explored by the multiple time scales method. Then nonlinear finite element analysis (FEA) is performed to verify the accuracy of the lumped stiffnesses of the diaphragm. The microphone is characterized and nonlinear least-squares technique is implemented to identify system parameters from experimental data. Finally uncertainty analysis is performed. The experimentally identified natural frequency and nonlinear stiffness parameter fall into their theoretical ranges for a 95% confidence level respectively.