Table 3 Composition of the features array utilized in the machine learning, ANN, and CNN 1D classifiers after the feature extraction process.
From: A dataset for environmental sound recognition in embedded systems for autonomous vehicles
Feature | Coefficients | Mean | Mean, std | Mean, std, median, skewness, kurtosis | ∑ |
|---|---|---|---|---|---|
RMS | 1 | 1 | — | — | 1 |
ZCR | 1 | 1 | — | — | 1 |
SC | 1 | 1 | — | — | 1 |
SB | 1 | 1 | — | — | 1 |
SRP | 1 | 1 | — | — | 1 |
Mel | 128 | 1 | — | — | 128 |
Δ MFCC | 13 | — | 2 | — | 26 |
ΔΔ MFCC | 13 | — | 2 | — | 26 |
MFCC | 13 | — | — | 5 | 65 |
SCT | 7 | — | — | 5 | 35 |
Chroma | 12 | — | — | 5 | 60 |
Tonnetz | 6 | — | — | 5 | 30 |
Total | 375 | ||||
