Complexity / 2020 / Article / Tab 2

Research Article

Articulatory-to-Acoustic Conversion Using BiLSTM-CNN Word-Attention-Based Method

Table 2

The RMSE (Hz) and r on the test set when using GMM, BiLSTM-RNN, and BiLSTM-CNN for EMA-to-F2 conversion.

GMMBiLSTM-RNNBiLSTM-CNN

RMSE37.1326.4222.10
r0.5650.6250.738

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.