For a numerical model of vibro-acoustic coupling analysis, such as a vehicle noise and vibration, both structural and acoustical dynamic characteristics are necessary to replicate the physical phenomenon. The accuracy of the analysis is not enough for substituting a prototype phase with a digital phase in the product development phases. One of the reasons is the difficulty of addressing the interior acoustical characteristics due to the complexity of the acoustical transfer paths, which are a duct and a small hole of trim parts in a vehicle. Those complex features affect on the nodal locations and the body coupling surface of acoustic mode shapes. In order to improve the accuracy of the analysis, the physical mechanisms of those features need to be extracted from experimental testing.The accuracy of the vibro-acoustic coupled system model for the low frequency range depends on how accurately modal characteristics are represented at the input, output, and the structure-acoustic coupling surface. Therefore, this study focus on extracting the detailed acoustic mode shapes on the coupling surface for the improvement of the model accuracy. The non-linear least square method as the one utilized in the previous study was applied to the new test data sets of an actual vehicle.In the previous study, it had one remaining issue, which was how to extract acoustic mode shapes in the frequency range of higher damping and higher acoustic modal densities. In order to solve this issue, the number of acoustic excitation was increased considering acoustic mode shapes. The eight loudspeakers were utilized as an acoustical excitation to excite acoustic modes evenly in the acoustic interior dimensions for higher frequency.With the results of this testing, the acoustic modes of an actual vehicle with heavy damping were accurately extracted as the complex mode shapes of no phase lag between nodes, which looks similar the un-coupled normal modes without rotation in animation up to 200Hz. The synthesized FRFs were replicated well with only the extracted several dominant acoustic mode shapes.