Missions

This position is in the field of natural language and speech processing. The successful candidate will join the international EUREKA BuddyBeWell project. He/she will implement multimodal machine learning approaches with LLMs for emotion detection from audio and text. He/she will be able to use an initial corpus of human-robot interactions collected in Woz and annotated on emotions. He/she will have to complete this first corpus with a second collection of data from elderly people. The protocol will be submitted to the Saclay University ethics committee. A protocol for validating interactions with Buddy will also be written and submitted to the ethics committee. A prototype will be built as part of the project. An initial phase of testing will be carried out internally at LISN. We will try to have a second phase of continuous testing in a home for the elderly. Research into the ethical dimensions of these interactions will be carried out.

Activities

- Participate in the project and collaborate with the project's various research teams.
- Collect data in Woz and supervise emotion annotations in a residence for the elderly.
- Build a multimodal audio and text-based emotion detection system for elderly voices.
- Test Buddy by following the validation protocol internally at LISN, then if possible externally in an elderly people's home.
- Assess the ethical dimensions of Human-Robot interactions

Skills

- Master's degree in computer science, automatic language processing or similar.
- Skills in supervised and semi-supervised machine learning, including deep learning and LLMs.
- Experience in automatic language processing
- Good command of English, both written and spoken
- Ability to work independently and as part of a team
- Ability to prioritize tasks and take initiative.

Work Context

This position is part of the international EUREKA BuddyBeWell project funded in France by BPI. This robotics project aims to develop ways of helping people age in good health. Many negative effects on (mental) health can be mitigated by lifestyle interventions that promote active, positive and social living. Social robotics could be the key to effective implementation of these interventions. Evidence from recent literature reviews on the impact of social robotics on the psychological well-being of older people is encouraging. The aim of this project is to provide elderly users (and their carers) with a robotic companion featuring the latest artificial intelligence technologies. The robot will provide personalized advice on healthy living, enabling people to stay at home longer.

The position is located in a sector under the protection of scientific and technical potential (PPST), and therefore requires, in accordance with the regulations, that your arrival is authorized by the competent authority of the MESR.

Constraints and risks

RAS

Additional Information

-