Analysing Sound Characteristics of Cello and Violin Using Fast Fourier Transform

The unique sound characteristics of music are based on multiple harmonic frequencies that exist within the sound waves. Through Fast Fourier Transform (FFT) software, the wave can be broken down into frequency and amplitude components. Spectrum analysis can be used quantitatively to describe these sound characteristics. In this paper, the frequency range present in the spectrum and the average intensity of the first 10 high notes in the sound are used to classify the sound characteristics of the cello and violin. This is done by generating a frequency (x-axis) and amplitude (y-axis) graph for the sounds of the cello and violin. The frequency and amplitudes are used to calculate 7 descriptors for sound characteristics, namely the centroid (f̅) Affinity (A), Brightness or Sharpness (S), Harmonicity (H), Monotony (M), Mean Affinity (MA), and Mean Contrast (MC). The results of the research reveal that quantitative frequency data analysis can generate and map sound characteristics. Quantitative analysis allows the quality of sound characteristics to be transformed into information understood by the computer. Cello string A3 is the most affinity (having a minimum A value, approaching 1, indicating that fo and f̅ are close). Cello string A3 and violin string D4 have the brightest sound (maximum S value). Violin strings D4 and E5 are the most harmonic (having a minimum H value, approaching 0). Cello string A3 shows harmonics present in most successive reductions (negative M) after fo. The MA value of cello C2 indicates dense secondary sound and is close to f̅. The maximum MC value of violin string D4 indicates normalized amplitude for its high-frequency secondary. The spectral analysis through FFT software convert sound quality characteristics into alternative sensory signals, which provides an opportunity for the deaf to experience sound through a different sensory experience.