Predicting Pronunciations with Syllabification and Stress with Recurrent Neural Networks

  Abstract