Table 5 The DM statistics of ten separate EFS evaluations for the original and synthesized features
From: An evolutionary feature synthesis approach for content-based audio retrieval
FS | Basic database (6 classes) | Extended database (12 classes) | ||
|---|---|---|---|---|
Original DM | Synthesized DM (μ ± σ) | Original DM | Synthesized DM (μ ± σ) | |
Segment | ||||
STAT | 500.5 | 43.6 ± 6.7 | 1793 | 406.3 ± 47.6 |
MFCC | 333.4 | 61.5 ± 8.5 | 1141 | 361.0 ± 22.4 |
Δ-MFCC | 515.7 | 130.5 ± 7.9 | 1865 | 620.1 ± 27.1 |
ΔΔ-MFCC | 622.6 | 140.9 ± 11.5 | 2072 | 532.7 ± 23.6 |
LPC | 1114 | 270.3 ± 8.9 | 4520 | 1204 ± 29.1 |
LPCC | 1638 | 310.3 ± 12.8 | 10830 | 1395 ± 26.5 |
S_AUDIO | 342.6 | 56.0 ± 4.1 | 1365 | 382.8 ± 15.2 |
Key-frame | ||||
MFCC + deltas | 2627 | 820.6 ± 53.6 | 7113 | 3093 ± 138 |
LPC + LPCC | 6951 | 2143 ± 39.4 | 23 900 | 6378 ± 157 |
K_AUDIO | 3150 | 782.7 ± 50.5 | 10 140 | 2774 ± 154 |