Table 1 Information on mixed data
From: Music detection from broadcast contents using convolutional neural networks with a Mel-scale kernel
Mixed data | Data 1 | Data 2 | Target dB (k) | Total duration (h) |
|---|---|---|---|---|
Music with speech | Library music | Librivox | − 30–0 dB | 25 |
Music with noise | Library music | ESC-50 | 0–30 dB | 25 |
Speech with noise | Librivox | ESC-50 | 0–30 dB | 25 |