DNN + Dropout

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio2 results

Speech Recognitiononswb_hub_500 WER fullSWBCH
Percentage error· 2014-06-30
19.1
best: 6.8 (IBM (LSTM+Conformer encoder-decoder))
SOTA
Building DNN Acoustic Models for Large Vocabulary Speech Recognition arXiv:1406.7806
Speech RecognitiononSwitchboard + Hub500
Percentage error· 2014-06-30
15
best: 4.3 (IBM (LSTM+Conformer encoder-decoder))
SOTA
Building DNN Acoustic Models for Large Vocabulary Speech Recognition arXiv:1406.7806