Accuracy of Automatic Processing of Speech-Language Pathologist and Child Talk During School-Based Therapy Sessions.

PURPOSE: This study examines the accuracy of Interaction Detection in Early Childhood Settings (IDEAS), a program that automatically transcribes audio files and estimates linguistic units relevant to speech-language therapy, including part-of-speech units that represent features of language complexity, such as adjectives and coordinating conjunctions. METHOD: Forty-five video-recorded speech-language therapy sessions involving 27 speech-language pathologists (SLPs) and 56 children were used. The F measure determines the accuracy of IDEAS diarization (i.e., speech segmentation and speaker classification). Two additional evaluation metrics, namely, median absolute relative error and correlation, indicate the accuracy of IDEAS for the estimation of linguistic units as compared with two conditions, namely, Oracle (manual diarization) and Voice Type Classifier (existing diarizer with acceptable accuracy). RESULTS: The high F measure for SLP talk data suggests high accuracy of IDEAS diarization for SLP talk but less so for child talk. These differences are reflected in the accuracy of IDEAS linguistic unit estimates. IDEAS median absolute relative error and correlation values for nine of the 10 SLP linguistic unit estimates meet the accuracy criteria, but none of the child linguistic unit estimates meet these criteria. The type of linguistic units also affects IDEAS accuracy. CONCLUSIONS: IDEAS was tailored to educational settings to automatically convert audio recordings into text and to provide linguistic unit estimates in speech-language therapy sessions and classroom settings. Although not perfect, IDEAS is reliable in automatically capturing and returning linguistic units, especially in SLP talk, that are relevant in research and practice. The tool offers a way to automatically measure SLP talk in clinical settings, which will support research seeking to understand how SLP talk influences children's language growth.

The preschool classroom linguistic environment: Children's first-person experiences.

The linguistic environment of the classroom is influential to young children's language development. To date, however, literature on the linguistic environment of child-care centers has largely examined teacher practices or children's aggregate environment, overlooking the child's first-person experiences and differentiated experiences within the classroom. In this study we used a new method in the educational setting that captures the learner's perspective: head-mounted cameras. Thirteen children in one preschool classroom wore a head-mounted camera to capture their first-person experiences in one morning session, including interactions with others and the features of the child-directed speech (CDS) addressed to them. Results revealed that, from children's personal view, the linguistic environment of the classroom is more dynamic from what previous studies have reported. Children interacted for longer with their teachers than their peers and heard more CDS from them, but for some children peers served as an additional source of language. Further, our analysis highlighted within-classroom variability in language experiences in terms of the properties of the CDS addressed to target children and how they were exposed to this input over time. Results are discussed with respect to peer influence on children's learning, heterogeneity in learning opportunities in classrooms, and the variability of the linguistic environment over time.

When Children Are Not Read to at Home: The Million Word Gap.

OBJECTIVE: In the United States, there are numerous ongoing efforts to remedy the Word Gap: massive differences in heard vocabulary for poor versus advantaged children during the first 5 years of life. One potentially important resource for vocabulary exposure is children's book reading sessions, which are more lexically diverse than standard caregiver-child conversations and have demonstrated significant correlational and causal influences on children's vocabulary development. Yet, nationally representative data suggest that around 25% of caregivers never read with their children. METHOD: This study uses data from 60 commonly read children's books to estimate the number of words that children are exposed to during book reading sessions. We estimated the total cumulative word exposure for children who are read to at varying frequencies corresponding to nationally representative benchmarks across the first 5 years of life. RESULTS: Parents who read 1 picture book with their children every day provide their children with exposure to an estimated 78,000 words each a year. Cumulatively, over the 5 years before kindergarten entry, we estimate that children from literacy-rich homes hear a cumulative 1.4 million more words during storybook reading than children who are never read to. CONCLUSION: Home-based shared book reading represents an important resource for closing the Word Gap.