A Systematic Review of Online Speech Therapy Systems for Intervention in Childhood Speech Communication Disorders

Abstract

Currently, not all children that need speech therapy have access to a therapist. With the current international shortage of speech–language pathologists (SLPs), there is a demand for online tools to support SLPs with their daily tasks. Several online speech therapy (OST) systems have been designed and proposed in the literature; however, the implementation of these systems is lacking. The technical knowledge that is needed to use these programs is a challenge for SLPs. There has been limited effort to systematically identify, analyze and report the findings of prior studies. We provide the results of an extensive literature review of OST systems for childhood speech communication disorders. We systematically review OST systems that can be used in clinical settings or from home as part of a treatment program for children with speech communication disorders. Our search strategy found 4481 papers, of which 35 were identified as focusing on speech therapy programs for speech communication disorders. The features of these programs were examined, and the main findings are extracted and presented. Our analysis indicates that most systems which are designed mainly to support the SLPs adopt and use supervised machine learning approaches that are either desktop-based or mobile-phone-based applications. Our findings reveal that speech therapy systems can provide important benefits for childhood speech. A collaboration between computer programmers and SLPs can contribute to implementing useful automated programs, leading to more children having access to good speech therapy.

Exploring the Role of Machine Learning in Diagnosing and Treating Speech Disorders: A Systematic Literature Review

“…It is clear to notice that the majority of proposed SLR focused on only one type of speech disorder, such as 32 and, 33 where only aphasia and Dysarthria are studied, or on a specific type of language, such as Tonal Languages in. 74 Other SLRs 34,35 pay attention to one patient's age. In, 36 the focus is on assistive technologies used as assessment technology for speech disorder patients.…”

“… 56 The learning set is used to build the machine learning model, while the testing set is used to evaluate the final model’s performance and generalization. 35 We need to process and turn the user’s speech into a set of features to use ML algorithms.…”

“…Most proposed SLRs focused on only one type of speech disorder, such as, 32,33 where only aphasia and Dysarthria are studied, respectively. Other SLRs 34,35 pay attention to one patient's age: children's age. In, 36 the focus is on assistive technologies used.…”

A look at phonetics and the pragmatic component of language from a speech therapy point of view

“…(10) Se concibe como una oportunidad de estimular el lenguaje y la comunicación en general, movilizar los procesos psicológicos, sensoriales, afectivos, interpersonales, cognitivos con el fin de contribuir al desarrollo del aprendizaje y favorecer las habilidades lingüísticas y comunicativas. (22) La atención logopédica posee sus particularidades y sus componentes básicos, sustentados en lo (teóricopráctico) y de las relaciones estructurales y jerárquicas de dicha atención; se enmarcan los siguientes: (23,24,25) • Prevenir los trastornos del lenguaje y la comunicación.…”

Improving Text-Independent Forced Alignment to Support Speech-Language Pathologists with Phonetic Transcription

“…Automatic speech recognition (ASR) is a technology that enables the conversion of spoken language into written text, making use of machine learning algorithms and acoustic models [ 17 , 18 ]. Over the years, significant advancements in neural networks, such as recurrent neural network (RNN) [ 19 ], bi-directional long short-term memory (BLSTM) [ 20 ], connectionist temporal classification (CTC) [ 21 ], and variants based on the generic networks, have been instrumental in advancing ASR, particularly from the 1990s to the 2010s [ 22 ].…”