Table 7 The retrieval performances obtained with the original features for both audio databases
From: An evolutionary feature synthesis approach for content-based audio retrieval
FS | Basic database (6 classes) | Extended database (12 classes) | ||
|---|---|---|---|---|
ANMRR | AP (%) | ANMRR | AP (%) | |
Segment | ||||
STAT | 0.358 | 61.0 | 0.514 | 45.9 |
MFCC | 0.228 | 73.8 | 0.367 | 60.2 |
Δ-MFCC | 0.351 | 62.2 | 0.507 | 46.6 |
ΔΔ-MFCC | 0.379 | 59.5 | 0.531 | 44.4 |
LPC | 0.549 | 43.1 | 0.686 | 29.8 |
LPCC | 0.569 | 41.1 | 0.717 | 26.9 |
S_AUDIO | 0.263 | 71.0 | 0.450 | 52.3 |
Key-frame | ||||
MFCC + deltas | 0.460 | 51.4 | 0.568 | 41.0 |
LPC + LPCC | 0.637 | 34.8 | 0.787 | 20.4 |
K_AUDIO | 0.375 | 59.5 | 0.524 | 44.9 |