Efficacy of Visual-Acoustic Biofeedback Intervention for Residual Rhotic Errors: A Single-Subject Randomization Study.
Tara McAllister Byun
https://pubmed.ncbi.nlm.nih.gov/28389677/Actively Recruiting
Led by New York University · Updated on 2026-05-13
76
Participants Needed
2
Research Sites
26 weeks
Total Duration
Speech Sound Disorder SSD affects many school-aged children, leading to social and emotional challenges that may continue into adolescence and adulthood. Researchers are studying the effects of an AI-guided home practice program to support speech sound generalization after participants complete a related biofeedback treatment study called VISIT. This study aims to see if using AI tools at home can improve speech sound accuracy in children and adolescents with SSD. Participants in this study will be randomized to receive either treatment as usual TAU or treatment as usual plus an AI-guided maintenance program with biofeedback TAUAI. The AI-guided practice uses a web interface called staRt that provides real-time visual feedback of speech sounds and classifier feedback on the accuracy of the American English r sound. All treatment sessions happen at home, while evaluations before and after the six-week GAINS study occur in the lab. During the study, participants will be assessed on speech sound accuracy by blinded listeners who rate their production of r sounds in word probes before and after the treatment period. Researchers will also collect data on regular community-based care during the study. The study starts with a six-week treatment phase, with continuous monitoring of progress and speech accuracy. Participants must have completed the VISIT study and meet specific language, hearing, and computer access requirements to take part.
CONDITIONS
Generalization With AI Navigation Using StaRt
T
Tara McAllister, PhD
T
Twylah Campbell, MA
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here
New York University
Lead Sponsor
Syracuse University
Collaborating Sponsor
This table compares participation in this trial with other options available for this condition. The information is drawn from published research and the trial’s study details and is provided for educational purposes only.
| Factors | staRt AI-Guided Visual-Acoustic Biofeedback | Community-Based Speech Therapy or No Additional Treatment |
|---|---|---|
| What it is | A home-based training program for speech sound disorder. It uses the staRt web-based software for practicing 'r' sounds, featuring real-time visual-acoustic biofeedback and AI feedback from the PERCEPT classifier. Sessions are facilitated but not directly led by clinicians. | Participants continue with their typical care outside the trial. This often involves school-based or privately delivered speech therapy, or, for some, no additional structured treatment. |
| Role in study | Serves as the investigational arm, evaluating whether structured, AI-supported practice using staRt biofeedback promotes correct 'r' sound production at home. | Functions as the comparator arm. Participants are monitored while receiving usual community care or no further structured intervention. |
| Primary use or goal | To determine if technology-assisted, AI-guided practice with immediate biofeedback can improve carryover of correct 'r' sounds in everyday speech outside therapy sessions. | Provides a real-world reference by assessing outcomes with continuation of routine speech therapy or no additional treatment outside of structured research context. |
| How it is given or performed | Participants use a web interface at home, practicing speech targets presented by the staRt software. Real-time acoustic biofeedback and periodic AI-based feedback are provided. Research assistants facilitate sessions but do not offer direct clinical guidance. | Participants attend community-based therapy sessions (such as school or private practice) or may receive no structured treatment, per their usual access to services. |
| Current evidence status | Preliminary studies, including small pilot and single-subject designs, suggest that visual-acoustic biofeedback (with or without AI support) may help some children improve production of challenging speech sounds like /r/. However, larger randomized controlled trials are needed, and conclusive evidence for staRt and integrated AI guidance is not yet established. | Community-based speech therapy is the established standard for treating speech sound disorders in school-age children. Outcomes are variable; some children achieve generalization of correct sounds, while others require alternative approaches. Evidence base is larger than for novel biofeedback interventions. |
| Important limitations | Evidence for staRt and AI-guided home-based biofeedback is still emerging, with limited large-scale data. Technology use requires access to computer hardware and reliable internet, may have usability barriers, and outcomes may depend on participant engagement. Some children may respond better than others. | Community-based therapy access may be inconsistent across participants, which can impact treatment intensity and generalization. The quality and amount of therapy received in routine settings is variable and difficult to control or standardize. |
Disclaimer: This comparison is created with the help of artificial intelligence(AI) and may contain errors or missing information. It is not medical advice and does not mean one option is better or safer than another. Clinical trials are still being studied, and joining a trial is your choice. Please consult your healthcare professional before making any decisions.
You may qualify if you...
You will not qualify if you...
Duration - 2 to 4 weeks
Participants are screened for eligibility to participate in the trial.
1 visit (in-person)
Duration - 6 weeks
Participants complete either AI-guided maintenance training with visual-acoustic biofeedback at home or follow treatment as usual in their community setting.
Home-based sessions using online tools with occasional research assistant support; 2 laboratory visits for pre- and post-treatment evaluations
Study Type
INTERVENTIONAL
Masking
SINGLE
Allocation
RANDOMIZED
Model
PARALLEL
Primary Purpose
TREATMENT
Number of Arms
2
Have more questions? Get in touch with our team for quick support
Tara McAllister Byun
https://pubmed.ncbi.nlm.nih.gov/28389677/Tara McAllister Byun, Peter F Halpin, Daniel Szeredi
https://pubmed.ncbi.nlm.nih.gov/25578293/Laura C Ochs, Megan C Leece, Jonathan L Preston...
https://pubmed.ncbi.nlm.nih.gov/38764857/Nina R Benway, Jonathan L Preston, Elaine Hitchcock...
https://pubmed.ncbi.nlm.nih.gov/37319018/Elaine R Hitchcock, Tara McAllister Byun
https://pubmed.ncbi.nlm.nih.gov/25216375/Tara McAllister Byun, Elaine R Hitchcock, Michelle T Swartz
https://pubmed.ncbi.nlm.nih.gov/25088034/Tara McAllister Byun, Heather Campbell
https://pubmed.ncbi.nlm.nih.gov/27891084/Tara McAllister Byun, Elaine R Hitchcock
https://pubmed.ncbi.nlm.nih.gov/22442281/Laura Peterson, Christian Savarese, Twylah Campbell...
https://pubmed.ncbi.nlm.nih.gov/35050705/New York University
Lead Sponsor
Syracuse University
Collaborating Sponsor