Research Article
Discrete and Dual Tree Wavelet Features for Real-Time Speech/Music Discrimination
Table 4
Average computation times for used feature extraction methods.
| | | Speech (msn.) | Music (msn.) |
| | Conventional | Common features | 0.2768 | 0.2745 |
| DWT based
Energy | Daubechies8-based energy features | 0.0216 | 0.0217 | | Coiflet1-based energy features | 0.0176 | 0.0176 |
| | DWT based | Haar | 0.0357 | 0.0382 | | Daubechies2 | 0.0401 | 0.04 | | Daubechies8 | 0.0485 | 0.0462 | | Daubechies15 | 0.1035 | 0.1034 | | Daubechies20 | 0.155 | 0.1547 |
| | CWT based | -shift-based CWT features | 0.0298 | 0.0296 | | CFS-based CWT features | 0.0301 | 0.03 |
|
|
