Differentiable Pooling for Unsupervised Acoustic Model Adaptation
Comparing a High and Low-level Deep Neural Network Implementation for Automatic Speech Recognition