Researchers have developed a method to track the convergence of children's speech patterns towards adult models using self-supervised speech representations. They analyzed longform recordings of children who are deaf or hard-of-hearing and their female caregivers, finding that the distance between the two groups' speech patterns decreases as the child gets older. This method uses a single metric to assess spoken language development and has been related to multiple standardized measures of speech and language in infancy through preschoolhood. --- Why it matters: This work matters because it provides a scalable and language-neutral way to assess spoken language development, which could be useful for researchers and clinicians working with children who are deaf or hard-of-hearing. It also opens up possibilities for more efficient and effective evaluation of language development across different populations and languages. Source: https://arxiv.org/abs/2608.20396