Table 1 Overall accuracy (in %) for direct classification into speech, instrumental and song
From: A hierarchical approach for speech-instrumental-song classification
| Â | T1 | T2 | ||||
|---|---|---|---|---|---|---|
Classification | Audio | MFCC | A+B | Audio | MFCC | A+B |
scheme | texture | (B) | Â | texture | (B) | Â |
| Â | (A) | Â | Â | (A) | Â | Â |
K-means | 54.00 | 52.67 | 54.88 | 51.19 | 50.00 | 52.67 |
MLP | 62.00 | 71.78 | 60.00 | 60.68 | 69.14 | 59.20 |
RANSAC | 79.77 | 72.22 | 79.33 | 78.88 | 70.18 | 79.08 |