How do characteristics of caregiver speech contribute to a child's early word learning? We explore the relationship between a single child's vocabulary growth and the distributional and prosodic characteristics of the speech he hears using data collected for the Human Speechome Project, an ecologically valid corpus collected from the home of a family with a young child. We measured F0, intensity, phoneme duration, usage frequency, recurrence, and MLU for caregivers' production of each word that the child learned during the period of recording. When all variables are considered, we obtain a model of word acquisition as a function of caregiver input speech. Coefficient estimates in the model help to illuminate which factors are relevant to learning classes of words. In addition, words that deviate from the model's prediction are of interest as they may suggest important social, contextual and other cues relevant to word learning.