Li-GRU

Reported on 4 benchmarks 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.

Audio4 results

Speech RecognitiononDIRHA English WSJ
Word Error Rate (WER)· 2018-11-19
23.9
SOTA
The PyTorch-Kaldi Speech Recognition Toolkit arXiv:1811.07453
Speech RecognitiononTIMIT
Percentage error· 2018-11-19
16.3
best: 8.3 (wav2vec 2.0)
The PyTorch-Kaldi Speech Recognition Toolkit arXiv:1811.07453
Speech RecognitiononLibriSpeech test-clean
Word Error Rate (WER)· 2018-11-19
6.2
best: 0.985 (United Med ASR)
The PyTorch-Kaldi Speech Recognition Toolkit arXiv:1811.07453
Speech RecognitiononCHiME real
Percentage error· 2018-11-19
14.6
best: 7.8 (Distortion-Independent + WRBN + Utterance-Wise Recurrent Dropout + Magnitude Features)
The PyTorch-Kaldi Speech Recognition Toolkit arXiv:1811.07453