Phoneme Recognition on TIMIT core (test)
10.3PERTrain from oracle
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Train from oracleGreedy decoding=true2024.02 | 10.3 | — | |
| REBORNGreedy decoding=true2024.02 | 13.5 | — | |
| EUROGreedy decoding=false, Language Model=4-gram, Backbone=WavLM2024.02 | 14.6 | — | |
| HMM over Time and Frequency Convolutional Net2015.06 | 16.7 | — | |
| Baseline + Conv. Features + Smooth FocusConvolutional Attention Features=true, Smooth Focus=true2015.06 | 17.6 | — | |
| PRETRANS-3L-250HWeights=4.3M, Training Method=Pretrained RNN Transducer, Number of Layers=3, Hidden Units=250, Cell Type=LSTM, Bidirectional=true2013.03 | 17.7 | 144 | |
| RNN Transducer2015.06 | 17.7 | — | |
| wav2vec-U+WFSTGreedy decoding=false, Language Model=4-gram2024.02 | 17.8 | — | |
| Baseline + Conv. FeaturesConvolutional Attention Features=true2015.06 | 18 | — | |
| TRANS-3L-250HWeights=4.3M, Training Method=RNN Transducer, Number of Layers=3, Hidden Units=250, Cell Type=LSTM, Bidirectional=true2013.03 | 18.3 | 112 | |
| CTC-5L-250HWeights=6.8M, Cell Type=LSTM, Number of Layers=5, Hidden Units=250, Bidirectional=true2013.03 | 18.4 | 150 | |
| CTC-3L-250HWeights=3.8M, Cell Type=LSTM, Number of Layers=3, Hidden Units=250, Bidirectional=true2013.03 | 18.6 | 124 | |
| Baseline Model2015.06 | 18.7 | — | |
| CTC-3L-421H-UNIWeights=3.8M, Cell Type=LSTM, Number of Layers=3, Hidden Units=421, Bidirectional=false2013.03 | 19.6 | 115 | |
| EUROGreedy decoding=false, Language Model=4-gram, Backbone=wav2vec 2.02024.02 | 19.8 | — | |
| CTC-2L-250HWeights=2.3M, Cell Type=LSTM, Number of Layers=2, Hidden Units=250, Bidirectional=true2013.03 | 21 | 55 | |
| wav2vec-U (reproduced)Greedy decoding=true2024.02 | 22.2 | — | |
| CTC-1L-622HWeights=3.8M, Cell Type=LSTM, Number of Layers=1, Hidden Units=622, Bidirectional=true2013.03 | 23 | 87 | |
| CTC-1L-250HWeights=0.8M, Cell Type=LSTM, Number of Layers=1, Hidden Units=250, Bidirectional=true2013.03 | 23.9 | 82 | |
| CTC-3L-500H-TANHWeights=3.7M, Cell Type=tanh, Number of Layers=3, Hidden Units=500, Bidirectional=true2013.03 | 37.6 | 107 |