Journal of Big Data

Table 3 Classification accuracies, F-measures and RTF of error detection obtained with the four models trained using the best feature set

From: Evaluation of the effectiveness and efficiency of state-of-the-art features and models for automatic speech recognition error detection

Feature set	Accuracy (%)	F-measure (%)			RTF
Feature set	Accuracy (%)	Correct	Error	Average	RTF
MLP	85.17	90.82	61.42	84.66	\(0.29 \times 10^{-4}\)
ULSTM	84.67	90.61	58.13	83.80	\(2.97 \times 10^{-4}\)
BLSTM	85.19	90.80	61.93	84.75	\(7.94 \times 10^{-4}\)
V-RNN	86.58	91.67	65.42	86.17	\(0.28 \times 10^{-4}\)

Back to article page