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

Table 1

Architectures of BLSTM, seq2seq, and seq2seq+attention.

Type	Type of layer(s)	Number of layer(s)	Number of units
BLSTM	BLSTM	2	256
seq2seq	LSTM	1	256
	BLSTM	2	256
seq2seq+attention	LSTM	1	256
	BLSTM	2	256