Now published in Neuropsychologia in the special issue in honor of John Duncan! 🎉  authors.elsevier.com/c/1l0HB6TBG5nQO

Humans largely learn language through speech. In contrast, most LLMs learn from pre-tokenized text. In our #Interspeech2025 paper, we introduce AuriStream: a simple, causal model that learns phoneme, word & semantic information from speech. Poster P6, Aug 19 at 13:30, Foyer 2.2!