This project aims to develop a personalized mobile health application for voice training using community-based participatory research, human-computer interaction, and personal health informatics methods. Voice training is a complex clinical practice that involves collaboration with a speech therapist to determine an individual?s needs and goals. It typically consists of personalized sessions that support individuals in changing their voices (such as modifying pitch, resonance, or speech patterns). The reasons why people may pursue voice training are varied, but often includes discomfort with voice being misaligned with gender identity. People may experience pressure to conform to a certain type of voice, and training may be inaccessible due to health disparities in particular minority populations. The proposed research will work within these constraints to design a novel voice training program to be deployed via a smartphone application. In doing so, the project will make several unique contributions to scientific understanding. No published research thus far has explored how data visualization techniques can be employed to communicate meaningful information about speech with a health- related end goal. Furthermore, this project will attempt to present these data insights through a mobile application, with the goal of delivering personalized, actionable health information directly to individuals. This will be accomplished by completing two primary aims. First, an application will be designed in partnership with community members and organizations using a community-based participatory research methodology. These methods have a rich history in public health; their goal is to engage at-risk communities in the research process from start to finish. Participants will assist with the development of the application such that it reflects the true needs of users. These community relationships will also build technological capacity and promote interest in community health and research. Second, the efficacy of the application will be evaluated using a within-subjects design. Outcome measures will be obtained using both qualitative (in-depth semi-structured interviews) and quantitative research methods (questionnaire responses, fundamental frequency, and resonance as measured by vowel formants). These data will be analyzed with the goal of typifying individual change over a 3 month period, during which participants will spend half the time interacting with the app.
Aim 2 will thus establish the app?s feasibility and usability, uncover inroads for future updates and improvements, as well as provide data to inform exploratory hypotheses regarding its efficacy. As a whole, my proposed project will advance the NIDCD?s mission to conduct community-engaged, health-promoting research via the development of assistive technology, particularly for disadvantaged populations. I am committed to completing this research via the training plan I have developed with my sponsors, whose mentorship will be extremely valuable as I work toward my goal of being an independent academic investigator.
Nationwide surveys reveal that lack of access to healthcare, limitations in provider knowledge, and denials of insurance coverage all negatively and disproportionately impact the health of individuals from minority groups in the United States. Personal health technologies can empower individuals to take control of their health and well-being through tailored interventions. The present research will contribute to our understanding of voice training interventions and investigate how interactive systems can deliver usable speech data, analyses, and feedback to users on a mobile device platform to support healthy behaviors, particularly in the context of a disadvantaged population.