Multilingual AI Agent Assistants (MAIA)

Type: National Project

Duration: from 2020 Apr 01 to 2023 Apr 01

Financed by: ANI e FCT

Prime Contractor: Unbabel Research (Company)

Project Web Site:

On MAIA, we’ll develop a multilingual conversational platform where human agents are assisted by AI agents. This approach will overcome the above limitations by targeting the following scientific and technological goals: New memory efficient neural models for context-aware machine translation, suit- able for online and real-time translation. These models will retain key aspects of a conversation (e.g., the gender of the customer), bringing them up whenever needed to translate a message. New answer generation techniques where the human agent (e.g., a tourism officer) will receive suggestions that reduce effort and increase the customer’s (e.g. a tourist) satisfaction. New techniques for conversational quality estimation and sentiment analysis to assess how well the conversation is addressing the customer’s needs, while simultaneously increasing --human empathy. Integration of the scientific advances above into a full end-to-end product. Two demonstrators will be built to cover concrete use cases in the Travel and Tourism industries.


  • Carnegie Mellon University (University) - Pittsburgh, USA
  • INESC-ID (Other)
  • IT (Other)
  • Unbabel (Company) - Lisboa, Portugal

Principal Investigators