Summary
William Havard is a postdoctoral researcher in computational linguistics with nine years of experience building end-to-end speech and NLP solutions for low-resource and endangered languages, currently based at Université d'Orléans. He combines fieldwork — including multi-week recording campaigns in Mauritius and Bolivia — with large-scale ML training (wav2vec2, HuBERT, CPC) to create speech processing tools for French-based Creoles and understudied communities. His work spans data engineering for terabyte-scale long-form corpora, unsupervised lexical discovery, and MLOps-driven model deployment, supervised student projects, and interdisciplinary collaboration with speech therapists. Trained to PhD level (Université Grenoble Alpes) and fluent across computational and linguistic methods, he brings rare hands-on experience bridging field linguistics and state-of-the-art self-supervised learning.
9 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD, Computational Linguistics, Ph.D., Doctor of Philosophy - PhD, Computational Linguistics, Ph.D. at Université Grenoble Alpes
Bachelor's degree, German Language and Literature, Bachelor's degree, German Language and Literature at Université Rennes 2
French, English, German, Spanish, creoles and pidgins, french-based