Table 4 Comparison of two-factor and...

Table 4

Comparison of two-factor and three-factor disentanglement on singing voices.

Factors			MSE ↓	Phoneme estimation [%]			Pitch estimation [%]
Timbre	Variation	Pitch	MSE ↓	(Timbre ↓)	Variation ↑	Pitch ↓	(Timbre ↓)	Variation ↓	Pitch ↑
✓		✓	0.873	(N/A)	—	64.7	(N/A)	—	16.6
	✓	✓	0.653	(—)	62.1	46.3	(—)	13.3	44.2
✓	✓	✓	0.613	(N/A)	61.4	48.1	(N/A)	12.7	45.5

Factors			MSE ↓	Phoneme estimation [%]			Pitch estimation [%]
Timbre	Variation	Pitch	MSE ↓	(Timbre ↓)	Variation ↑	Pitch ↓	(Timbre ↓)	Variation ↓	Pitch ↑
✓		✓	0.873	(N/A)	—	64.7	(N/A)	—	16.6
	✓	✓	0.653	(—)	62.1	46.3	(—)	13.3	44.2
✓	✓	✓	0.613	(N/A)	61.4	48.1	(N/A)	12.7	45.5