Associate professor (Maître de conférences)
The associate professor position is composed of a teaching and researching position.
Teaching at the University of Lorraine (in the computer sciences department of Faculté des Sciences et Technologies – Nancy):
in « Licence d’informatique » and « Master d’informatique »
- Speech Analysis, Processing and Synthesis (M1, M2)
- Systems (process synchronization…) (L2, L3)
- Computer systems (Linux, Shells…)
- Computer tools (C2i, NUMOC) (L1)
- IA introduction (M2 Law IP/IT)
Research at the LORIA (in MULTISPEECH team)
- My research topics are:
- Speech synthesis
- Expressive speech synthesis (IA – Deep Learning)
- acoustic by information transfer (emotion, style…)
- audiovisual (face, gesture ANR Syncogest 2025-2029 )
- Text-To-Speech synthesis by HMM and DNN (CMCU project PHC Utique [2015-2018] ).
- Acoustic-Visual Speech Synthesis (ViSAC Project [ANR] 2009-2012).
- Text-To-Speech synthesis by unit selection (Soja-tts)
- Expressive speech synthesis (IA – Deep Learning)
- Disentanglement of attributes, anonymization (Coordinator of ANR SpeechPrivacy 2024-2028)
- Speech analysis/synthesis techniques.
- Speech synthesis
- keywords : speech, synthesis, Text-to-Speech, expressive speech, anonymization, disentanglement, (time/scale modifications, foreign language acquisition, prosody, oral comprehension)
Current administrative charge at the University of Lorraine
Co-Responsable/Correspondant Numoc UL (outils et culture numériques)
Software
JCorpusRecorder
JCorpusRecoder is a software for the recording of audio corpora. It allows the recording in a « Studio » mode for a free recording or in a perceptive experience mode (listening constraint or number of repetitions). The user interface is in French, English or German (the documentation is in French and English). It is an audio recording tool that is easy to use for the person recording and fully customizable. This software has been used primarily in a research context in language learning but also in other contexts. [More here, for details and download]
Background
Researcher in a postdoctoral position at the MULTITEL research center (2002/2003).
I worked for the TTS synthesis group to add Non-Uniform Unit selection in eLite project : a Linguistically-Oriented Non-uniform units Selector.
Thesis : « Speech analysis/synthesis techniques applied to language learning »
I defended my dissertation in 24th september 2002
Research team
The MULTISPEECH team focuses its research on spoken communication. It covers a wide range of activities including the study of human perception of acoustic clues, the analysis of speech and automatic recognition. These works come within the scope of building robust vocal interfaces in multimodal and embedded systems. Acoustic to Articulatory Mapping, Acoustical and Perceptual Cues for Speech Sounds, Stochastic Models for Automatic Speech Recognition, Speech Synthesis…
