Abstract
This research introduces an innovative methodology for evaluating and predicting soundboard quality in the intricate craftsmanship of sape instruments. Despite the sape’s profound cultural significance, the process of selecting soundboard wood has been inadequately explored, resulting in uncertainties within the crafting community. Addressing this research gap, this study integrates advanced machine learning techniques and devises a specialized Graphical User Interface (GUI) tailored for sape makers. The methodology encompasses a thorough acoustic analysis of three distinct hardwoods—adau, merbau, and tapang—employing machine learning classification through Support Vector Machine with a Gaussian kernel. The study culminates in the development of a user-friendly GUI for soundboard quality assessment. Results underscore the model’s proficiency for achieving an optimized accuracy of 87.8% in classifying sape audio samples. The MATLAB App Designer-based GUI streamlines the evaluation process, offering a practical and accessible tool for craftsmen. This integrated approach, harmonizing traditional craftsmanship with cutting-edge technology, holds the potential to revolutionize sape instrument manufacturing, ensuring the preservation and progressive evolution of this rich cultural heritage.