Samples of Utterance-level Sequential Modeling For Deep Gaussian Process Based
Speech Synthesis Using Simple Recurrent Unit

utteranceLSTM-RNNSRU-NNSRU-BayesNNFF-DGPSRU-DGP (Proposed)
BASIC5000_0068
BASIC5000_0168
BASIC5000_0313
BASIC5000_0647
BASIC5000_1300
BASIC5000_1616


MethodsMeaning
LSTM-RNNConventional NN-based method using LSTM blocks.
SRU-NNConventional NN-based method using SRU blocks.
SRU-BayesNNBayesien NN-based method using SRU blocks.
FF-DGPConventional DGP-based method using feedforward blocks.
SRU-DGPProposed DGP-based method using SRU blocks.