Deep multimodal learning for Audio-Visual Speech Recognition | IEEE Conference Publication | IEEE Xplore