Comparison of two types of latent representations obtained using small and large encoders during re-entry training on monophonic musical fragments.
| Encoder | MSE ↓ | Instrument classification [%] | Pitch estimation [%] | ||||
|---|---|---|---|---|---|---|---|
| Timbre ↑ | Variation ↓ | Pitch ↓ | (Timbre ↓) | Variation ↓ | Pitch ↑ | ||
| Small | 0.431 | 51.2 | 19.3 | 37.5 | (N/A) | 11.1 | 79.3 |
| Large | 0.443 | 51.8 | 30.7 | 40.5 | (N/A) | 23.9 | 82.0 |
| Encoder | MSE ↓ | Instrument classification [%] | Pitch estimation [%] | ||||
|---|---|---|---|---|---|---|---|
| Timbre ↑ | Variation ↓ | Pitch ↓ | (Timbre ↓) | Variation ↓ | Pitch ↑ | ||
| Small | 0.431 | 51.2 | 19.3 | 37.5 | (N/A) | 11.1 | 79.3 |
| Large | 0.443 | 51.8 | 30.7 | 40.5 | (N/A) | 23.9 | 82.0 |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.