4 ms·Residual LSTM: Deep Recurrent Architecture for Distant Speech Recognition2 points by brandonb 10y ago