Job openings: PhD studentship in Formal Linguistics/ Socially Interactive Agents/Machine learning,
Université Paris Cité, Sorbonne Université, IRCAM
The application deadline is June 27th, 2023
Applications to be sent to: yonatan.ginzburg (at) u-paris.fr AND you must also apply through the official site:
https://emploi.cnrs.fr/Offres/Doctorant/UMR7110-SOPLEM-011/Default.aspx?lang=EN
We are inviting applications for one PhD position (3 years) (funding available 10/2023-09/2026) in the area of formal linguistics, Socially Interactive Agents (SIAs), and machine learning.
The positions will be funded as part of the CNRS 80-PRIME The Laughing Infant project jointly held by Jonathan Ginzburg (Laboratoire de Linguistique Formelle (LLF), Université Paris Cité) , Catherine Pelachaud (L'institut des systèmes intelligents et de robotique (ISIR), Sorbonne Université), and Nicolas Obin (Lab Sciences et Technologies de la Musique et du Son, IRCAM).
The goal of The Laughing Infant (TELIN) project is to synthesize work on language acquisition, on the semantics and pragmatics of laughter, on Socially Interactive Agents (SIAs), on analysis and synthesis of laughter, and on machine learning, with the aim of developing an SIA who will emulate an infant during the acquisition of laughter and by using this SIA to test several learning algorithms that take as input different modalities (audio, facial expression, language) and contexts (playing with toys, interacting with parents and siblings) and output laughter.
The PhD topic is to develop formal and computational models that compute when and how a baby robot (the robot Furhat with the mask of a baby) responds to a participant’s expression and activity. The focus is on the production of laughter in infants. This involves (i) Analysis of a corpus of baby laughter (ii) Development of a rigorous theoretical analysis of laughter during an interaction between a carer and infants, (iii the development of a computational model based on deep learning that simulates when laughter should be triggered. The models will be evaluated objectively as well as through experimental studies.
The successful candidate will work in three areas: formal linguistics, virtual agent development, and machine learning. They must have a master's degree in the fields of linguistics, computer science or cognitive science.
The expected skills are: -At least two of the following three skills: (i) Strong expertise in formal semantics or pragmatics (ii) Strong expertise in machine learning, and in particular in deep neural networks.
(iii) Proficiency in python and java.
In addition: - Excellent command of spoken and written scientific English
- Autonomy, teamwork, productivity, rigor and methodology
The successful candidate will be accommodated part of the time at LLF, part of the time at ISIR and part of the time at IRCAM, in all cases with a personal workstation. They will benefit from access to the FurHat robot and the team's GPU computing servers, as well as the servers of the Jean Zay supercomputer hosted by CNRS-IDRIS. They will have access to the Ircam media library, as well as to the CNRS and Sorbonne University online libraries.
Applicants are requested to submit their application, including a cover letter that specifies why you would like to work on this topic and what qualifies you for it, an academic CV, your MSc thesis (or a current draft), copies of academic degree certificates, and names of two potential references.
The applications should be sent via email directly to Prof. Jonathan Ginzburg: yonatan.ginzburg (at) u-paris.fr
The application deadline is June 27th, 2023
Laboratoire de Linguistique Formelle (LLF) (Université Paris Cité, http://www.llf.cnrs.fr), L'institut des systèmes intelligents et de robotique (ISIR) (Sorbonne Université, https://www.isir.upmc.fr) and The Sciences et Technologies de la Musique et du Son lab (IRCAM and Sorbonne Université, https://www.stms-lab.fr/team/analyse-et-synthese-des-sons) are internationally prominent CNRS labs in their domains.
—LLF has strengths in formal linguistics (ranging from phonetics and phonology to
semantics, pragmatics, and the analysis of multimodal dialogue, and interfacing with sociolinguistics), psycholinguistics (on spoken and signed languages using EEG, EGG, and Eye Trackers) , and computational linguistics (in particular using a variety of deep learning techniques).
—ISIR research focuses on robotics and AI. Gathered in multidisciplinary teams, researchers create drones, micro-tweezers, bionic prostheses, social robots, surgical arms and all kinds of intelligent and interactive systems, physical, virtual or mixed reality. ISIR maintains one of the largest robot fleets in Europe.
—The Sciences et Technologies de la Musique et du Son lab at IRCAM participates in the renewal of musical expression through the contributions of computer science, acoustics, signal processing, cognitive sciences, and musicology. The focus is on music and organized sound in a unique context where contemporary creation meets scientific and technological research. In particular, the Sound Analysis & Synthesis team carries out research and development activity in sound analysis, transformation, and synthesis of sound signals.