COURSE OVERVIEW

 This practical and highly-interactive course includes real-life case studies and exercises where participants will be engaged in a series of interactive small groups and class workshops.  

This course is designed to provide participants with a detailed and up-to-date overview of Machine Translation. It covers the scope of machine translation (MT) and the differences between human and machine translation; the types of machine translation approaches and linguistic challenges in machine translation; the rule-based machine translation (RBMT), statistical machine translation (SMT), neural machine translation (NMT) and encoder-decoder architecture in NMT; the attention mechanism in NMT and transformer models in machine translation; and the pre-trained language models for translation. 

 

Further, the course will also discuss the data collection and preprocessing for MT, training large-scale MT models and evaluating machine translation output; handling low-resource languages in MT and the bias and ethical issues in machine translation; the domain adaptation for machine translation, real-time and low-latency machine translation and multimodal machine translation (MMT); the speech-to-speech machine translation and federated and privacy-preserving MT models; and the Export MT Models for Deployment (ONNX, TensorFlow Lite) and run machine translation in a web application. 

During this interactive course, participants will learn the API-based translation using google translate and deepL; the difference between cloud versus edge deployment for MT services; the AI-powered human-machine collaboration in translation and neural-symbolic hybrid approaches in MT; the AI’s role in preserving endangered languages and advancements in AI-powered speech translation; how AI helps human translators improve accuracy; and the automated post-editing for faster translation workflows and AI in game and website localization.