Psychoacoustic Music Analysis Based on the Discrete Wavelet Packet Transform

Table 1

Power spectrum portion under masking threshold [5].

Audio type	PE model (%)	Proposed DWPT-based model (%)	Gain (%)
Country	54	73	19
Jazz	50	74	24
Pop	58	74	16
Rock	72	77	5
Classic	44	78	34
AVERAGE	55	75	20