Due to constant evolution in both noise regulations and noise comfort standards, noise reduction inside the vehicle remains one of the main issues faced today by the automotive industry. One of the most efficient methods for noise reduction is the introduction of acoustic treatments, made of multilayered trimmed panels. Constraints on these components, such as weight, packaging space and overall sound quality as well as the amount of possible material and geometrical combinations, have led automotive OEMs to use innovative methods, such as numerical acoustic simulation, so as to evaluate noise transmission in a fast and cost-effective way. While the computational cost for performing such analyses is insignificant for a limited number of configurations, the evaluation of multiple design parameter combinations early in the design stage can lead to non-viable computation times in an industrial context. This paper presents a framework for the efficient, almost real-time evaluation of quality indicators, such as the sound transmission loss, using machine learning techniques, with data from a limited amount of vibroacoustic simulations. The method is evaluated on several firewall panels covering a large design space, where the sound transmission loss of the panels can be predicted with good accuracy across the frequency spectrum. Furthermore, the method is applied to the design space covering the properties of individual materials with similar outputs. The resulting models can be further used for optimizing the behavior of the acoustic treatment. The performance of the proposed methodology is demonstrated on an industrial firewall panel application.