Speech Recognition with Deep Recurrent Neural Networks

Benchmark Model Rank Results
speech-recognition-on-timitBi-LSTM + skip connections w/ CTC#15Percentage error: 17.7