The classical guitar is a popular string instrument whose sound is created through a coupled mechanical process. Typically, the different parts of a guitar are made of different types of wood. The variability of the used woods’ material parameters causes the vibrational behavior of even seemingly identical instruments to vary, and thus each instrument develops an individual sound. In this contribution, a method is presented that allows the non-destructive identification of the most influential material parameters of a complete guitar. First, the modal parameters of a specific guitar are determined by an experimental modal analysis. The guitar is then modeled using CT scans and discretized for the finite element method. In addition to the detailed geometry, the model includes the orthotropic material parameters of the different wood species and the fluid-structure interaction between the body and the enclosed air. This model will subsequently be used to identify the material parameters of the real guitar. For this purpose, many evaluations of the parameterized model are necessary, which leads to very long computation times. Hence, parametric model order reduction is applied, which can significantly reduce the model order and thus the computational time, while maintaining the parameter dependence in the reduced model.