Automated analyses of full-day recordings were used to determine whether young children who are hard-of-hearing (HH) received similar levels of exposure to adult words and conversational interactions as age-matched peers with normal-hearing (NH). Differences in adult input between children in this study and in a normative database were considered. Finally, factors were examined that may have contributed to individual differences in the input characteristics of families. Results indicated that the NH and HH groups were exposed to similar numbers of adult words and conversational turns. However, both the NH and HH groups were exposed to more adult words and engaged in more conversational turns than the NH children in the normative sample. Considering only the HH group, both quantity of adult words and conversational exchanges were correlated with children's auditory characteristics. Children's receptive language ability was correlated with conversational exchanges but not with adult word counts.

Objectives The objectives of this study were to examine the quantity of adult words, adult-child conversational turns, and electronic media in the auditory environments of toddlers who are hard of hearing (HH) and to examine whether these variables contributed to variability in children’s communication outcomes. Design Participants were 28 children with mild to severe hearing loss. Full-day recordings of children’s auditory environments were collected within 6 months of their 2nd birthdays by utilizing LENA (Language ENvironment Analysis) technology. The system analyzes full-day acoustic recordings, yielding estimates of the quantity of adult words, conversational turns, and electronic media exposure in the recordings. Children’s communication outcomes were assessed via the receptive and expressive scales of the Mullen Scales of Early Learning at 2 years of age and the Comprehensive Assessment of Spoken Language at 3 years of age. Results On average, the HH toddlers were exposed to approximately 1400 adult words per hour and participated in approximately 60 conversational turns per hour. An average of 8% of each recording was classified as electronic media. However, there was considerable within-group variability on all three measures. Frequency of conversational turns, but not adult words, was positively associated with children’s communication outcomes at 2 and 3 years of age. Amount of electronic media exposure was negatively associated with 2-year-old receptive language abilities; however, regression results indicate that the relationship was fully mediated by the quantity of conversational turns. Conclusions HH toddlers who were engaged in more conversational turns demonstrated stronger linguistic outcomes than HH toddlers who were engaged in fewer conversational turns. The frequency of these interactions was found to be decreased in households with high rates of electronic media exposure. Optimal language-learning environments for HH toddlers include frequent linguistic interactions between parents and children. To support this goal, parents should be encouraged to reduce their children’s exposure to electronic media.

Automatic speech processing (ASP) has recently been applied to very large datasets of naturalistically collected, daylong recordings of child speech via an audio recorder worn by young children. The system developed by the LENA Research Foundation analyzes children's speech for research and clinical purposes, with special focus on of identifying and tagging family speech dynamics and the at-home acoustic environment from the auditory perspective of the child. A primary issue for researchers, clinicians, and families using the Language ENvironment Analysis (LENA) system is to what degree the segment labels are valid. This classification study evaluates the performance of the computer ASP output against 23 trained human judges who made about 53,000 judgements of classification of segments tagged by the LENA ASP. Results indicate performance consistent with modern ASP such as those using HMM methods, with acoustic characteristics of fundamental frequency and segment duration most important for both human and machine classifications. Results are likely to be important for interpreting and improving ASP output.

HomeBank is introduced here. It is a public, permanent, extensible, online database of daylong audio recorded in naturalistic environments. HomeBank serves two primary purposes. First, it is a repository for raw audio and associated files: one database requires special permissions, and another redacted database allows unrestricted public access. Associated files include metadata such as participant demographics and clinical diagnostics, automated annotations, and human-generated transcriptions and annotations. Many recordings use the child-perspective LENA recorders (LENA Research Foundation, Boulder, Colorado, United States), but various recordings and metadata can be accommodated. The HomeBank database can have both vetted and unvetted recordings, with different levels of accessibility. Additionally, HomeBank is an open repository for processing and analysis tools for HomeBank or similar data sets. HomeBank is flexible for users and contributors, making primary data available to researchers, especially those in child development, linguistics, and audio engineering. HomeBank facilitates researchers’ access to large-scale data and tools, linking the acoustic, auditory, and linguistic characteristics of children’s environments with a variety of variables including socioeconomic status, family characteristics, language trajectories, and disorders. Automated processing applied to daylong home audio recordings is now becoming widely used in early intervention initiatives, helping parents to provide richer speech input to at-risk children.