Alice Coucke
Research
You can find my publications on my Google Scholar Profile.
Current
For Sonos Voice Control, we consider the problem of automatic speech recognition on the edge, e.g. on small devices typical of IoT applications, with privacy in mind. I filed 3 patents currently under review. We also released several open evaluation datasets for spoken language understanding and keyword spotting that have now become standard benchmarks in the field. I have worked on several topics: fairness in speech technologies, speech-to-text systems, keyword spotting in the context of wake word detectors, natural language understanding, sentence generation, etc. Over the years, I maintained a strong interest in the scientific questions raised when assessing the performance of a machine learning model in realistic conditions.
- Sonos Voice Control Bias Assessment Dataset: A Methodology for Demographic Bias Assessment in Voice Assistants (pre-print)
- C. Sekkat, F. Leroy, S. Mdhaffar, B.P. Smith, Y. Estève, J. Dureau, and A. Coucke
- May 2024, The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING)
- Small Footprint Text-Independent Speaker Verification For Embedded Systems
- J. Balian, R. Tavarone, M. Poumeyrol, and A. Coucke
- April 2021, International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2021, poster
- Conditioned Text Generation with Transfer for Closed-Domain Dialogue Systems
- S. D'Ascoli, A. Coucke, F. Caltagirone, A. Caulier, and M. Lelarge
- November 2020, International Conference on Statistical Language and Speech Processing 2019, oral presentation
- Conditioned Query Generation for Task-Oriented Dialogue Systems
- S. D'Ascoli, A. Coucke, F. Caltagirone, A. Caulier, and M. Lelargeo
- November 2019, arXiv preprint
- Spoken Language Understanding at the Edge
- A. Saade, A. Coucke, A. Caulier, J. Dureau, A. Ball, T. Blüche, D. Leroy, C. Doumouro, T. Gisselbrecht, F. Caltagirone, T. Lavril, and M. Primet
- October 2019, Energy Efficient Machine Learning and Cognitive Computing workshop NeurIPS 2019, poster and oral presentation
- Efficient Keyword Spotting using Dilated Convolutions and Gating
- A. Coucke, M. Chlieh, T. Gisselbrecht, D. Leroy, M. Poumeyrol, and T. Lavril
- November 2018, International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2019, poster
- Federated Learning for Keyword Spotting
- D. Leroy, A. Coucke, T. Lavril, T. Gisselbrecht, and J. Dureau
- October 2018, International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2019, poster
- Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces
- A. Coucke, A. Saade, A. Ball, T. Blüche, A. Caulier, D. Leroy, C. Doumouro, T. Gisselbrecht, F. Caltagirone, T. Lavril, M. Primet, and J. Dureau
- May 2018, Privacy in Machine Learning and Artificial Intelligence worskhop at ICML 2018, spotlight presentation
PhD
I pursued my PhD under the supervision of Rémi Monasson and Martin Weigt at the lab of theoretical physics of the École Normale Supérieure (Paris). I focused on maximum entropy graphical model inference and its application to genomic data, especially protein structure prediction. I discussed the extension of these approaches to other challenging fields, such as sequence folding prediction and homology detection. Through an extensive study on both artificial and biological data, I provided a better interpretation of the central inferred parameters, up to then poorly understood. I presented a new and more precise procedure for the inference of generative models, which lead to further improvements on real, finitely sampled data.
- Inference of compressed Potts graphical models
- F. Rizzato, A. Coucke, E. De Leonardis, J. Barton, J. Tubiana, R. Monasson, and S. Cocco
- July 2019, Physical Review E
- Direct coevolutionary couplings reflect biophysical residue interactions in proteins
- A. Coucke, G. Uguzzoni, F. Oteri, S. Cocco, R. Monasson, and M. Weigt
- November 2016, The Journal of chemical physics
- ACE: adaptive cluster expansion for maximum entropy graphical model inference
- J. Barton, E. De Leonardis, A. Coucke, and S. Cocco
- May 2016, Bioinformatics
- Statistical modeling of protein sequences beyond structural prediction: High-dimensional inference with correlated data
- A. Coucke
- October 2016, PhD thesis
Misc
In January 2018, I took part in the DAT-ICU datathon for intensive care, aiming at presenting a clinical project using the MIMIC database (data about 50 000 intensive care unit patients). We proposed to associate a clinical print to each patient stay, allowing for instance to identify clusters of similar patients in a non supervised approach. We presented an algorithm of clinical data (textual and numerical) dimensionality reduction based on deep learning techniques. My team won the first prize.
- Deep representation for patient visits from electronic health records
- J. Escudié, A. Saade, A. Coucke, and M. Lelarge
- March 2018, arxiv preprint
During my master 2, I worked at the “Physico-Chimie” lab of the Institut Curie under the supervision of Jean-François Joanny. I developed a theoretical model based on active gel theory and performed numerical simulations to study the effect of cell migration on morphogenesis in biological tissues.
- An interplay of migratory and division forces as a generic mechanism for stem cell patterns
- E. Hannezo, A. Coucke, and JF. Joanny
- December 2015, Physical Review E