[Retracted] The Mongolian Vowel Acoustic Model Based on the Clustering Algorithm

Table 5

Invisible triphonon experiment results.

Subjects	Sentence level recognition experimental results				Word level recognition experimental results
	% Correct	H	S	N	% Correct	ACC	H	D	S	I	N
Mongol 1	9.67	58	542	600	77.67	68.45	9087	342	2271	1078	11700
Mongol 2	18.15	98	442	540	79.54	70.17	5727	137	1336	675	7200
Mongol-all	13.68	156	984	1140	78.38	69.11	14814	479	3607	1753	18900
Dialogue 1	16.03	469	2456	2925	63.55	45.71	8751	351	4668	2457	13770
Dialogue 2	18.66	435	1896	2331	60.33	43.35	6987	289	4305	1967	11581
dialogue 3	13.80	265	1655	1920	59.21	42.90	6590	376	4164	1815	11130
Dialogue-all	16.29	1169	6007	7176	61.20	44.10	22328	1016	13137	6239	36481