A novel voice conversion approach using admissible wavelet packet decomposition

EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Performance of proposed method for predicting formant frequencies within a specified percentage of deviation

Transformation model	Formant frequencies	% Predicted frame within deviation
		2%	10%	15%	20%	25%	50%	μ _RMSE	γ₍x, y)
M1-F1	f1	60	87	88	89	90	94	4.46	0.78
	f2	51	83	88	90	91	95	3.39	0.86
	f3	57	90	91	92	94	98	2.78	0.84
	f4	66	89	93	95	96	100	2.41	0.86
F2-M2	f1	35	65	71	76	80	89	3.36	0.71
	f2	44	88	92	93	94	97	2.71	0.86
	f3	58	89	91	94	95	99	2.40	0.74
	f4	81	90	94	96	97	100	2.10	0.77