After three and a half years spent completing an industrially funded PhD in Computer Science at A2iA and LIMSI-CNRS in Paris, France, on deep neural networks for handwriting recognition, I worked for 2 years as a Research Engineer at A2iA.

In June 2017, I joined Snips as a machine learning scientist to work on on-device automatic speech recognition systems.

I am now machine learning scientist at Sonos since its acquisition of Snips in November 2019, to help improve Sonos' voice experience.

For more information on my background, have a look at my resume.

I am mainly interested in innovation through machine learning techniques, with a particular focus on neural networks and deep learning approaches.

Site contents

Besides my resume, I use this website to present my research interests and a list of published papers, and to host my MSc dissertation on mathematical expression recognition and my PhD thesis on handwriting recognition with deep neural networks.

Articles / Posts

Scan, Attend and Read

Attention models for end-to-end handwritten paragraph recognition

Read more

CTC and the blank label

Read more