Table 1 Evaluation results for the tested methods using the balanced drums and accompaniment mixture as input.
From: Drum Sound Detection in Polyphonic Music with Hidden Markov Models
Method | Metric | BD | SD | HH | Total |
|---|---|---|---|---|---|
HMM |
| 84.7 | 65.3 | 84.9 | 80.0 |
| Â |
| 77.4 | 44.9 | 78.5 | 68.0 |
| Â |
| 80.9 | 53.2 | 81.6 | 73.5 |
HMM + MLLR |
| 80.2 | 66.3 | 84.7 | 79.0 |
| Â |
| 81.5 | 45.3 | 82.6 | 70.9 |
| Â |
| 80.8 | 53.9 | 83.6 | 74.7 |
HMM comb |
| 54.9 | 38.8 | 73.0 | 55.0 |
| Â |
| 66.4 | 47.0 | 58.7 | 57.4 |
| Â |
| 60.1 | 42.5 | 65.1 | 56.1 |
| Â |
| 69.9 | 57.0 | 58.2 | 62.0 |
NMF-PSA [11] |
| 57.9 | 16.7 | 53.5 | 43.6 |
| Â |
| 63.4 | 25.9 | 55.8 | 51.2 |
| Â |
| 80.9 | 65.9 | 47.1 | 54.3 |
SVM [6] |
| 38.4 | 14.2 | 69.5 | 43.8 |
| Â |
| 51.1 | 23.4 | 56.1 | 48.5 |
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
(%)