Deep and Wide: Multiple Layers in Automatic Speech Recognition | IEEE Journals & Magazine | IEEE Xplore