State-of-the-Art Speech Recognition Using Multi-Stream Self-Attention With Dilated 1D Convolutions

Open paper
Benchmark Model Rank Results
speech-recognition-on-librispeech-test-cleanMulti-Stream Self-Attention With Dilated 1D Convolutions#85Word Error Rate (WER): 2.20
speech-recognition-on-librispeech-test-otherMulti-Stream Self-Attention With Dilated 1D Convolutions#91Word Error Rate (WER): 5.80

Rank counts only results with a code link.