Smart speakers are an acceptable and feasible speech practice tool for children with speech difficulties

Abstract

Purpose: A record number of children are experiencing speech difficulties. This study explored whether mainstream smart speakers can support speech practice, and assesses a proposed five-factor model for how these improvements occur, based on: spaced learning, immediate feedback, autonomous motivation, reduced social barriers, and increased social interactions. Materials and methods: We conducted a mixed-methods feasibility study with 11 families of children with speech difficulties. Data included pre- and post-trial speech recordings (rated by 24 naïve raters), activity logs, post-trial surveys for parents and children, and parent interviews. We assessed device use, changes in speech clarity, alignment with the proposed 5-factor model, and barriers. Results: Children used smart speakers on average once daily for approximately 6 commands. Speech clarity ratings improved post-exposure and improvements correlated with device use. Parents noted children spoke more slowly and loudly to the device. Both parent and child reports reflected the 5-factor model, with children showing motivation, spaced engagement, and willingness to repeat commands. Conclusions: Naturalistic use of smart speakers is an acceptable and feasible tool to support speech practice in children with speech difficulties. Parents viewed smart speakers as a useful complement to formal therapy, though frustration when not understood remains a factor for implementation.