A proof-of-concept study for automatic speech recognition to transcribe AAC speakers’ speech from high-technology AAC systems
Reference
Chen, S. H. K., Saeli, C., & Hu, G. (2023). A proof-of-concept study for automatic speech recognition to transcribe AAC speakers’ speech from high-technology AAC systems. Assistive Technology, (just-accepted).
Abstract
Automatic speech recognition (ASR) is an emerging technology that has been used in recognizing non-typical speech of people with speech impairment and enhancing the language sample transcription process in communication sciences and disorders. However, the feasibility of using ASR for recognizing speech samples from high-tech Augmentative and Alternative Communication (AAC) systems has not been investigated. This proof-of-concept paper aims to investigate the feasibility of using AAC-ASR to transcribe language samples generated by high-tech AAC systems and compares the recognition accuracy of two published ASR models: CMU Sphinx and Google Speech-to-text. An AAC-ASR model was developed that transcribes simulated AAC speaker language samples. The AAC-ASR model’s word error rate (WER) was compared with those of CMU Sphinx and Google Speech-to-text. The WER of the AAC-ASR model outperformed (28.6%) compared with CMU Sphinx and Google when tested on the testing files (70.7% and 86.2% retrospectively). Our results demonstrate the feasibility of using the ASR model to automatically transcribe high-technology AAC-simulated language samples to support language sample analysis. Future steps will focus on developing the model with diverse AAC speech training datasets and understanding the speech patterns of individual AAC users to refine the AAC-ASR model.
Deep Neural Network-based Speaker-Aware Information Logging (SAIL) for Augmentative and Alternative Communication
Reference
Hu, G., Chen, S. H. K., & Mazur, N. (2021). Deep neural network-based speaker-aware information logging for augmentative and alternative communication. Journal of Artificial Intelligence and Technology, 1(2), 138-143.
Abstract
People with complex communication needs can use a high-technology augmentative and alternative communication device to communicate with others. Currently, researchers and clinicians often use data logging from high-tech augmentative and alternative communication devices to analyze augmentative and alternative communication user performance. However, existing automated data logging systems cannot differentiate the authorship of the data log when more than one user accesses the device. This issue reduces the validity of the data logs and increases the difficulties of performance analysis. Therefore, this paper presents a solution using a deep neural network-based visual analysis approach to process videos to detect different augmentative and alternative communication users in practice sessions. This approach has significant potential to improve the validity of data logs and ultimately to enhance augmentative and alternative communication outcome measures.
Augmentative and alternative communication intervention for in-patient individuals with post-stroke aphasia: Study protocol of a parallel-group, pragmatic randomized controlled trial
Reference
Huang, L., Chen, S. H. K., Xu, S., Wang, Y., Jin, X., Wan, P., ... & Shan, C. (2021). Augmentative and alternative communication intervention for in-patient individuals with post-stroke aphasia: Study protocol of a parallel-group, pragmatic randomized controlled trial. Trials, 22(1), 1-9.
Background
People with post-stroke aphasia commonly receive speech-language therapy (SLT) when they are admitted to hospitals. Commonly, these patients reported communication difficulties in in-patient settings. Augmentative and alternative communication (AAC) has been reported as an effective treatment approach to improve communication effectiveness, language performance, decreasing depression, and improving quality of life for this population. However, little evidence has demonstrated the use of AAC intervention (AACT) in early recovery from people with post-stroke aphasia in in-patient rehabilitation settings for improving these patients’ communication effectiveness. The pilot randomized controlled trial (RCT) will explore the effectiveness and feasibility of including AACT in regular SLT for in-patient people with post-stroke aphasia.
Method
This pilot RCT is a single-blind, randomized controlled trial with two parallel groups. Both groups receive a 1-h treatment session, including either both AACT and SLT or SLT only for ten consecutive days. We aim to include 22 in-patient participants with post-stroke aphasia in each group. Participants will be assessed at pre- and post-intervention and 2 weeks after intervention. The primary outcomes are the ability of communication measured by the communication of basic needs subtest in the Functional Assessment of Communication Skills for Adult (FACS) and the overall language performance measured by the Chinese Standard Aphasia Battery (ABC). The secondary outcomes include a 10-min conversation, the 10-item Hospital version of the Stroke Aphasic Depression Questionnaire (SADQH-10), the Stroke-Specific Quality of Life Scale (SS-QOL), and a patient and caregiver satisfaction questionnaire.
Discussion
This pilot RCT will contribute to new scientific evidence to the field of aphasia rehabilitation in early recovery during the in-patient period. The paper describes the trial, which will explore the effect of combining AACT and SLT and SLT only, our choice of primary and secondary outcome measures, and proposed analyses. The study results will provide information for implementing AACT in the regular in-patient SLT of future RCTs.
What Speech-Language Pathologists Should Know About Selecting Eye Gaze Augmentative and Alternative Communication Systems
Reference
Chen, S. H. K., & O'Leary, M. (2018). Eye gaze 101: What speech-language pathologists should know about selecting eye gaze augmentative and alternative communication systems. Perspectives of the ASHA Special Interest Groups, 3(12), 24-32.
Abstract
People with complex communication disabilities along with severe physical disabilities commonly need assistive technology to support access to augmentative and alternative communication (AAC) devices. Eye gaze techniques have become one of the solutions available to solve their access issues. An AAC system that can be used with eye gaze technology usually involves a computer-based device and an eye-tracking device. Although applying eye gaze as an alternative access method for AAC is promising for many people with both complex communication disabilities and physical disabilities, knowledge and skills of the clinician in gathering evidence to decide an eye gaze access is critical to achieve the desired outcome of effective communication. This article will review previous research evidence related to eye-tracking technologies and eye gaze techniques applied with different populations and, then, provide clinical guidance to readers.
The clinical trials mosaic: Toward a range of clinical trials designs to optimize evidence-based treatment
Reference
Ridenour, T. A., Chen, S. H. K., Liu, H. Y., Bobashev, G. V., Hill, K., & Cooper, R. (2017). The clinical trials mosaic: Toward a range of clinical trials designs to optimize evidence-based treatment. Journal for Person-Oriented Research, 3(1), 28.
Objective
Dichotomizing clinical trials designs into nomothetic (e.g., randomized clinical trials or RCTs) versus idiographic (e.g., N-of-1 orcase studies) precludes use of an array of hybrid designs and potential research questions between these extremes.This paper describes unique clinical evidence that can be garnered using idiographic clinical trials(ICTs) to complement RCT data.Proposedand illustrated herein is that innovative combinations of design features from RCTs and ICTs could provide clinicians with far more comprehensive information for testing treatments, conducting pragmatic trials, and making evidence-based clinical decisions.
Method
Mixed model trajectory analysis and unified structural equations modeling were coupled with multiple baseline designs in (a) a true N-of-1 pilot study to improve severe autism-related communication deficits and (b) a small sample preliminary study of two complimentary interventions to relieve wheelchair discomfort.
Results
Evidence supported certain mechanisms of treatment outcomes and ruled outothers.Effect sizes included mean phase differences (i.e., effectiveness), trajectory slopes, and differences in path coefficients between study phases.
Conclusion
ICTs can be analyzed with equivalent rigor as, and generate effect sizes comparable to, RCTs for the purpose of developing hybrid designs toaugment RCTs for pilot testing innovative treatment, efficacy research on rare diseases or other small populations, quantifyingwithin-person processes, and conductingclinical trials in many situations when RCTs are not feasible.