Five Short Pieces on Neural Machine Translation of Text and Speech - Lecture

Explore five cutting-edge research contributions in neural machine translation presented by Marcello Federico, Principal Applied Scientist at Amazon AI. Delve into recurrent neural MT for morphologically rich languages and transformer model enhancements for low-resource settings. Discover data augmentation techniques for integrating terminology handling and improving robustness against speech recognition errors. Learn about methods to bias NMT for producing shorter or longer translations. Gain insights from Federico's extensive experience in machine translation, spoken language translation, and language modeling throughout this informative 70-minute talk hosted by the Center for Language & Speech Processing at Johns Hopkins University.