Improving Fluency of Spoken Mandarin for Nonnative Speakers by Prosodic Boundary Prediction Based on Deep Learning

Table 4

Experimental results of different methods for PPH boundary prediction.

Model	Precision (%)	Recall (%)	score (%)
ME	71.77	77.24	74.40
CRF	78.61	81.55	80.05
BLSTM	87.84	74.76	80.77
BLSTM-CRF	—	—	82.95
seq2seq	83.40	81.76	82.57
seq2seq+attention+NSH	89.10	81.71	85.24