Microsoft 2016b

Reported on 1 benchmark across 1 task · 1 paper · 1 SOTA

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

Audio1 result

Speech RecognitiononSwitchboard + Hub500
Percentage error· 2016-10-17
5.8
best: 4.3 (IBM (LSTM+Conformer encoder-decoder))
SOTA
Achieving Human Parity in Conversational Speech Recognition arXiv:1610.05256