I am an AI Researcher at Vivoka where I lead the R&D on automatic speech recognition.

I have been a post-doctoral researcher in the COMPRISE project funded by European Union’s Horizon 2020 Research and Innovation Programme. My research focused on weakly supervised learning techniques to improve speech recognition in fully private-by-design conversational systems. I was working with the MultiSpeech group of the LORIA-INRIA laboratory in France and the Spoken Language Systems (LSV) group at Saarland University in Germany.

Earlier I was a researcher with the speech and NLP group at TCS Research and Innovation, where I contributed towards research and development on speech recognition and spoken language understanding for voice assistants, and several other speech and language processing applications.

I obtained a PhD from the Université de Lorraine in France. My PhD research explored Bayesian probabilistic models and continuous vector representations from neural networks to model topics and semantic context of out-of-vocabulary words in large vocabulary continuous speech recognition. I was part of the MultiSpeech group of the LORIA-INRIA laboratory in Nancy and the language processing team at LIA, Avignon Université, while I was funded by the ContNomina project grant of the French National Research Agency (ANR).