COMPARING SHORT-TIME FOURIER TRANSFORM (STFT) AND CONSTANT-Q TRANSFORM (QCT) FOR SPECTRAL ANALYSIS OF THE REBANA, MALAY TRADITIONAL MUSIC INSTRUMENT

Abstract

This study compares the Short-Time Fourier Transform (STFT) and Constant-Q Transform (CQT) techniques in capturing the intricate spectral characteristics of the rebana, a traditional Malay frame drum known for its rich harmonic content and transient attacks. High-quality audio recordings were obtained in a controlled studio environment using a Shure SM57 microphone placed six inches from the drumhead, ensuring minimal noise interference. The analyses were conducted in MATLAB utilizing the MIRToolbox, enabling robust computation of time-frequency representations and related spectral parameters. While STFT provides a relatively straightforward and computationally efficient method, its linear frequency resolution can limit the clarity of lower harmonic details, especially for instruments like the rebana that exhibit a broad range of tonal nuances. In contrast, CQT’s logarithmic frequency scaling aligns closely with human pitch perception, revealing more detailed harmonic structures and subtle timbral variations. Although this approach demands increased computational resources, the resulting representations offer valuable insights into the instrument’s acoustic complexity.  These findings suggest that CQT may be better suited for in-depth harmonic analysis and ethnomusicological research, whereas STFT remains a practical choice for applications emphasizing transient behavior or requiring real-time processing. Future work could explore hybrid strategies that combine the strengths of both transforms, yielding a more comprehensive toolkit for analyzing traditional percussion instruments and informing efforts in cultural preservation, pedagogy, and performance practice.