Apraxia World: An interactive, technology-based comprehensive tool for remote speech therapy of childhood apraxia of speech
- View All
Developmental disabilities and neuromotor disorders such as childhood apraxia of speech (CAS) can result in delays in speech articulation, language production and reading in children. Traditional CAS therapy involves extended one-on-one practice with a speech language pathologist (SLP), which can be difficult to manage due to time constraints and expensesâ . Thus, there is a need for practical and cost-effective technological interventions to complement traditional face-to-face therapy sessions. To address this need, in an earlier project the PIs developed a multi-tier system for remotely administering speech therapy to children with CAS. The system allows an SLP to create speech-production exercises and remotely assign them to each child, who can then practice them in the form of a basic game on a mobile device. The mobile app records the childâ s utterances and uploads them to a back-end server, where they are automatically scored by a speech-analysis engine. The SLP can then review the individual recordings and the automated scores through a web interface, provide feedback to the child, and adapt the training program as needed. The system has undergone testing with children diagnosed with CAS, their parents and SLPs. Results support the feasibility of the approach as a supplement to traditional face-to-face therapy, with the majority of children, parents and SLPs indicating preference towards our system compared to traditional therapy. However, feedback from study participants indicates that the mobile application lacks the degree of variety, interactivity and entertainment required to maintain the childâ s attention during a therapy session or promote long-term adherence to therapy programs. In addition, at present the application only supports a small set of the exercises in the speech therapy protocol. Using the existing system as a backbone, the specific aims of this proposal are to 1) develop interactive games that can be driven by the childâ s speech production, 2) develop speech-recognition algorithms on the tablet to provide real-time feedback, 3) improve our server-based scoring algorithms to meet the industry standard of 80% reliability against human judgments and handle complex speech productions; and 4) perform clinical studies to quantify the therapeutic efficacy of the system and its ability to promote adherence to regular practice outside the clinic.