While machine translation is now widely used in the language industry, applying MT to specific domains like the life scienes or movie subtitling remains a challenge. The data that is needed for domain-specific training purposes is limited and can be expensive to create. This webinar explores how knowledge from one domain can be transferred to another domain through the use of a pre-trained word vector model and neural network gates which control the flow of information between domains.
This webinar is part of the Research and Innovation Action "Quality Translation 21 (QT21)." This project has received funding from the European Union's Horizon 2020 program for ICT under grant agreement no. 645452.
Jian carried out his PhD research in Dublin City University, where he focused on domain adaptation in machine translation. In addition, his experiences also include software engineering, machine translation scientist, machine translation consultant, and language technology scientist. He works at the Centre for Digital Content Technology at Dublin City University and recently joined Voysis in Dublin, building on methods on speech synthesis.