. Author manuscript; available in PMC: 2016 Sep 1.

Published in final edited form as: IEEE/ACM Trans Audio Speech Lang Process. 2015 May 15;23(9):1421–1430. doi: 10.1109/TASLP.2015.2434213

TABLE V.

Comparative results on the dysarthric speech corpus

Method	Measure	Ataxic	ALS	HD	PD
Inten-Praat	SR error rate%	13.1	21.1	17.4	20.3
	SR mean error	0.35	0.41	0.50	1.06
	Mean error	2.21	3.32	2.76	3.30
TF-Corr	SR error rate%	15.7	25.2	19.7	16.8
	SR mean error	0.55	0.54	0.59	0.74
	Mean error	2.56	4.09	3.2	2.75
GMM-NN	SR error rate%	12.1	13.4	22.8	21.3
	SR mean error	0.31	0.31	0.65	1.07
	Mean error	2.10	2.20	3.87	3.48
OPT-Sent-Adap	SR error rate%	7.0	7.6	13.2	12.8
	SR mean error	0.23	0.21	0.42	0.69
	Mean error	1.20	1.30	2.26	2.12