Speech Prediction on TIMIT (test)
2.76MSELipschitz RNN
Evaluation results
| Method | Links | |
|---|---|---|
| Lipschitz RNNNumber of layers=1, Hidden dimension (N)=256, Number of parameters=≈198K, Integration scheme=RK22020.06 | 2.76 | |
| Lipschitz RNNNumber of layers=1, Hidden dimension (N)=256, Number of parameters=≈198K, Integration scheme=Euler2020.06 | 2.82 | |
| Exponential RNN (Lezcano-Casado & Martinez-Rubio, 2019)Number of layers=1, Hidden dimension (N)=425, Number of parameters=≈200K2020.06 | 5.48 | |
| SRLSTM (Nguyen et al., 2020)Number of layers=1, Hidden dimension (N)=158, Number of parameters=≈200K2020.06 | 5.83 | |
| MomentumLSTM (Nguyen et al., 2020)Number of layers=1, Hidden dimension (N)=158, Number of parameters=≈200K2020.06 | 5.87 | |
| Cayley RNN (Helfrich et al., 2018)Number of layers=1, Hidden dimension (N)=425, Number of parameters=≈200K2020.06 | 7.36 | |
| LSTM (Nguyen et al., 2020)Number of layers=1, Hidden dimension (N)=158, Number of parameters=≈200K2020.06 | 9.37 | |
| CConv-LSTM#PARAMS=≈ 88K2017.05 | 11.9 | |
| Conv-LSTM#PARAMS=≈ 88K2017.05 | 12.18 | |
| LSTM (Helfrich et al., 2018)Number of layers=1, Hidden dimension (N)=158, Number of parameters=≈200K2020.06 | 12.62 | |
| Full-capacity Unitary RNN (Wisdom et al., 2016)Number of layers=1, Hidden dimension (N)=256, Number of parameters=≈200K2020.06 | 14.45 | |
| Full-Capacity uRNN#PARAMS=≈ 135K2017.05 | 14.66 | |
| LSTM#PARAMS=≈ 135K2017.05 | 16.98 |