User Pic

Aghilas Sini

LST ( Le Mans )

(+33) +33 2 43 83 38 30




I am currently a teacher-researcher in the LST team at LIUM. Previously, I have been on contract at the University of Rennes in Lannion since October 2022. I obtained my Master’s degree in artificial intelligence from the Université Paul Sabatier in 2014. I then worked as a CNRS engineer at LORIA-INRIA in the Multispeech team in Nancy.

I received my doctorate in computer science from the University of Rennes 1 in 2020. I worked in the Expression team at IRISA (Lannion) on topics relating to speech synthesis (text-to-speech, voice conversion and voice cloning) and voice biometrics (in particular techniques for detecting voice impersonation, also known as deep fake speech). I have worked with the Rennes-based company Learn&Go on subjects related to automatic speech recognition for pronunciation analysis.

My research focuses on the automatic processing of expressive speech, more specifically in the fields of speech conversion, computer-assisted language acquisition and the protection of private data.

My research projects at LIUM are aimed at designing expressive multilingual multimodal speech translation tools.

HAL Publications  Hal[authIdHal_s][]=

LIUM Publications - 2

[1]Olivier Crouzet, Elisabeth Delais-Roussarie, Agnieszka Duniec, Loriane Leprieur, Patrick Louis Rohrer, et al.. Actes des XXXIVe Journées d'Études sur la Parole -- JEP 2022. ISCA; Nantes Université; Association Francophone de la Communication Parlée (AFCP), 2022, ⟨10.21437/JEP.2022⟩. <hal-03958018>
[2]Aghilas Sini, Damien Lolive, Gaëlle Vidal, Marie Tahon, Elisabeth Delais-Roussarie. SynPaFlex-Corpus: An Expressive French Audiobooks Corpus Dedicated to Expressive Speech Synthesis. Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), May 2018, Miyazaki, Japan. <hal-01826690>