A Simple Web Utility for Automatic Speech Quantification in Dyadic Reading Interactions

Adult-child reading interactions can produce efficient childhood learning outcomes when formal methods of adult-initiated prompting and dialog are employed. In order to quantify instances of such prompting, child development researchers often manually code speech recordings or transcripts taken from adult-child readings, which can be a laborious process. For researchers, it would therefore be valuable to have the capacity to automatically capture quantitative measures of communication from adult-child reading interactions. This paper introduces SoundCount, a new open-source utility for extracting descriptive features from recordings of speech during communication between an adult and child reading an eBook together. In a simple web-based framework, SoundCount consolidates functionalities for speech analysis and quantification in the context of dyadic reading interactions, specifically providing measures related to word count and speaker differentiation. Our preliminary results demonstrate the functional feasibility of SoundCount, and our technical discussion will enable readers to use SoundCount in their own research. Given the technical viability of SoundCount, future work will include the implementation of additional measures for the system, new features such as automated audio segmentation based on speaker, and a test of the efficacy of real-time feedback systems based on speech measures.