Artificial intelligence (AI) has demonstrated significant potential in supporting pediatric speech therapy.However, no systematic review has examined the quality and expanding landscape of commercially available AI-enabled speech therapy mobile apps for children.We conducted a systematic review and analysis of 21 identified commercially available AI-enabled apps designed for speech-sound practice with children.Using 15 evaluation criteria consolidated and extended from prior HCI design guidelines for human-AI and child-AI interaction, our content and risk analysis revealed that existing apps have not yet leveraged the full potential of existing AI tools, and often fail to rigorously adhere to the best recommended practice.Incorporating feedback from the speech therapy community, we categorized our findings into four areas of ethical concern.To address these concerns, we propose actionable recommendations for app designers, families, and speech therapists to guide more ethical development and safe use of AI-enabled applications for pediatric speech therapy.

Individuals with cognitive-communication disorders (CCDs) due to neurological conditions, such as traumatic brain injury and aphasia, experience difficulties in communication and cognition that impact their ability to perform activities of daily living, or ADLs (e.g., self-care, meal preparation, scheduling). Voice assistive technology (VAT) can support the independent performance of ADLs; however, there are limited VAT training programs that teach individuals with CCDs how to properly implement and use VAT for ADLs. The present study examined the implementation of an online training program using Alexa voice commands for five ADL domains (scheduling, entertainment, self-care, news & facts, and meal preparation). Using video analysis with seven adults with CCDs between ages 25 and 82 and interviews with five participants and three caregivers, we synthesized five weeks of training performance, analyzed participants' perceived benefits and challenges, and discussed challenges and opportunities for implementing VAT training for ADLs skills for adults with CCDs.

Children with high-functioning autism (HFA) often face challenges in emotional recognition and expression, leading to emotional distress and social difficulties. Conversational agents developed for HFA children in previous studies show limitations in children's learning effectiveness due to the conversational agents’ inability to dynamically generate personalized and contextual content. Recent advanced generative Artificial Intelligence techniques, with the capability to generate substantial diverse and high-quality texts and visual content, offer an opportunity for personalized assistance in emotional learning for HFA children. Based on the findings of our formative study, we integrated large language models and text-to-image models to develop a tool named EmoEden supporting children with HFA. Over a 22-day study involving six HFA children, it is observed that EmoEden effectively engaged children and improved their emotional recognition and expression abilities. Additionally, we identified the advantages and potential risks of applying generative AI to assist HFA children in emotional learning.

Due to challenges in communication and cognition (e.g., attention, memory, organization, etc.), individuals with cognitive-communication needs (CCNs) experience difficulties performing activities of daily living, or ADLs (e.g., self-care, meal preparation, managing different routines and appointments). Voice-assisted technology (VAT) has been shown to support these individuals for ADLs; however, there are limited clinician-guided VAT training programs available for individuals with CCNs to use VAT for ADLs. This study aims to design and implement a 6-week virtual VAT training program focused on using Amazon Alexa voice commands to improve independence in ADL tasks for three adults with CCNs. This research study describes the collaborative design process of the VAT Training program by a clinical team and the 6-week delivery of the training program. By analyzing training session videos and participant interviews, the benefits and challenges from the training, as well as future plans to improve the implementation of VAT for telehealth are discussed.

The popularity of automatic speech recognition (ASR) systems nowadays leads to an increasing need for improving their accessibility. Handling stuttering speech is an important feature for accessible ASR systems. To improve the accessibility of ASR systems for stutterers, we need to expose and analyze the failures of ASR systems on stuttering speech. The speech datasets recorded from stutterers are not diverse enough to expose most of the failures. Furthermore, these datasets lack ground truth information about the non-stuttered text, rendering them unsuitable as comprehensive test suites. Therefore, a methodology for generating stuttering speech as test inputs to test and analyze the performance of ASR systems is needed. However, generating valid test inputs in this scenario is challenging. The reason is that although the generated test inputs should mimic how stutterers speak, they should also be diverse enough to trigger more failures. To address the challenge, we propose Aster, a technique for automatically testing the accessibility of ASR systems. Aster can generate valid test cases by injecting five different types of stuttering. The generated test cases can both simulate realistic stuttering speech and expose failures in ASR systems. Moreover, Aster can further enhance the quality of the test cases with a multi-objective optimization-based seed updating algorithm. We implemented Aster as a framework and evaluated it on four open-source ASR models and three commercial ASR systems. We conduct a comprehensive evaluation of Aster and find that it significantly increases the word error rate, match error rate, and word information loss in the evaluated ASR systems. Additionally, our user study demonstrates that the generated stuttering audio is indistinguishable from real-world stuttering audio clips.

The rapid adoption of smart home speakers into households has enabled young children to verbally communicate with voice assistants (VAs). VAs, such as Amazon Alexa, allow children to use their voice to play various game-based activities and offer opportunities to understand children's communication during child-agent communication. This qualitative study describes findings from video analysis of young children between three and six years old who engaged in voice-based gameplay with two commercial Alexa Skills, Animal Sounds and Animal Game. We report findings about verbal behaviors and communication breakdowns and repairs between children and Alexa, and discuss key considerations for designing voice games for children.

The world is becoming increasingly multilingual. In the U.S., despite rapid growth in linguistic diversity, there is a complete lack of multilingual language assessment tools and a severe shortage of multilingual clinicians to detect language impairments among children who speak minority languages. To develop accessible child language assessment tools, we designed MECO-LAB, a web-based bilingual Mandarin-English assessment that uses parents as one of the potential groups of test administrators. We analyzed 16 videos of child-parent dyads and found that with minimal instructions, the majority (11 out of 16) of parents were capable of administering MECO-LAB to their children. We identified 296 interference and 381 support behaviors from parents that are influenced by linguistic, cognitive, emotional, and technical factors that researchers should consider when designing online language assessments. We proposed design recommendations for supporting child-parent interactions in similar applications that enable parents to administer online bilingual language assessments to their children.

To improve functional communication and behavioral management, many children with disabilities receive behavioral and communication-related intervention from professionals such as behavioral analysts and speech and language therapists. This paper presents user perspectives from three clinicians who have used and/or designed assistive technology with children with disabilities, and calls for researchers to recognize and leverage clinicians' knowledge to design accessible technology for children with complex sensory and communication needs.