The main project objectives are:
1. Research: Advancing research in responsible creation and application of multilingual, multimodal pretrained XR models
- Advance the state-of-the-art in multimodal, multilingual pre-trained models and their application to downstream tasks. This project will focus on making them customizable, efficient, and ethical.
2. Deployment: Development of new methods and algorithms for adapted, contextualised, and robust dialogue assistant XR models
- Leverage multimodal models to deliver tools which provide a new paradigm in quality, robustness, and contextualisation capabilities for dialogue assistant components such as transcription, translation, and summarisation.
3. Integration: Development of tools for online meetings and customer support agents
- Design and implementation of interfaces to visualise, edit and publish transcripts, minutes, combined with meeting platforms like Zoom to deliver multilingual transcription and translation support.
4. Exploitation: Sustainable, maintainable platform and services
- Development of a plan for sustainable exploitation and use of the platforms, systems and technologies developed in UTTER.
5. Dissemination: Dissemination and communication of project results to stakeholders
- Dissemination and communication of project results to stakeholders and user group. Successful promotion and use of platform in the pilots run under the financial support to third parties call.