In-loop operators for natural language interaction in advanced manufacturing environments

The main objective of the BERREKIN project is research and development in incremental learning technologies in conversational HMI in an OIL scenario so that systems can evolve and adapt to the context of each industrial plant, ensuring localised capture, storage and transfer of operators' expert knowledge.

The specific objectives of the project are as follows: • R&D of incremental learning techniques for dialogue systems, both statistical and based on ontologies and hybrid learning and exploitation policies, to provide greater flexibility to the systems.

• R&D of incremental learning techniques for question-answering or Q&A systems.

• R&D of the application of the OIL paradigm in distributed computing or Edge Computing environments for conversational HMI interactions.

• R&D on OIL interfaces that allow the operator to be involved in the circuit to adapt, customise and provide knowledge to conversational HMI systems.

• Compilation of resources for the Basque language to enable the implementation of conversational OIL & HMI systems in that language as well.
• Validation of a conversational HMI & OIL prototype integrated with real users.