Skip to main content
. 2021 Apr 29;4:648543. doi: 10.3389/frai.2021.648543

Table 5.

Human codes for Dataset B along with MFA's pronunciation assessment.

Human codes: N-N-N G-N-N G-G-N G-G-G Totals
MFA = -in 633 (90.3%) 94 (84.7%) 27 (61.4%) 69 (24.7%) 823 (72.5%)
MFA = -ing 68 (9.7%) 17 (15.3%) 17 (38.6%) 210 (75.3%) 312 (27.5%)
Totals 701 111 44 279

Shading indicates cells where human codes and MFA agree.