From: An evolutionary feature synthesis approach for content-based audio retrieval
Segment features | Key-frame features |
---|---|
STATa (39-D) | MFCC + Δ-MFCC + ΔΔ-MFCC (39-D) |
13 Mel-frequency cepstral coefficients (MFCC) (26-D) | 12th-order LPC + 14th-order LPCC (26-D) |
13 Δ-MFCC (26-D) | K_AUDIOc (31-D) |
13 ΔΔ-MFCC (26-D) | |
10th-order linear prediction coefficients (LPC) (20-D) | |
14th-order linear prediction cepstral coefficients (LPCC) (28-D) | |
S_AUDIOb (38-D) |