Skip to main content
. Author manuscript; available in PMC: 2012 Nov 27.
Published in final edited form as: J Acoust Soc Am. 1989 Aug;86(2):566–581. doi: 10.1121/1.398236

TABLE VIII.

MRT open-response phonemes accounting for the highest percentage of error in initial position.

System Phonemes
(Percentage of error accounted for in each evaluation)
Natural f
(29.41)
k
(23.50)
m
(11.76)
b
(11.76)
n
(05.88)
Prose 2000 V3.0 b
(24.79)
k
(22.59)
h
(12.40)
g
(08.82)
w
(06.06)
DECtalk Paul b
(42.90)
k
(16.30)
r
(16.30)
h
(06.00)
l
(06.00)
DECtalk Betty h
(23.70)
p
(22.20)
b
(16.30)
t
(07.40)
ð
(05.20)
MITalk-79 k
(19.50)
b
(17.30)
h
(13.10)
t
(11.70)
f
(06.70)
Amiga 500 p
(44.19)
h
(11.25)
k
(09.28)
b
(09.16)
t
(04.87)
Infovox f
(17.60)
w
(14.60)
b
(11.90)
s
(10.40)
h
(08.90)
Smoothtalker p
(18.88)
b
(17.58)
k
(12.29)
f
(06.99)
h
(06.51)
Votrax k
(15.39)
h
(10.75)
f
(08.41)
t
(06.32)
d
(05.30)
Echo p
(22.91)
b
(18.88)
k
(14.04)
h
(09.03)
d
(06.96)