An Automatic Error Detection Method for Engineering English Translation Based on the Deep Learning Model

Table 3

Experimental results of multitask mispronunciation detection and diagnosis model comparison.

Model	Sequence labeling accuracy	Accuracy	Correct
MT-D	66.64%	85.74%	88.93%
MT-D-P	69.94%	81 .95%	86.34%
MT-A	67.32%	86.22%	89.22%
MT-A-P	77.33%	87.34%	89.14%