Language sample analysis is a well-supported, evidence-based approach for analyzing discourse-level language abilities of individuals with a variety of communication disorders across the life-span. It has been used to quantify and compare discourse production of neurologically typical individuals to those with impairments that impact language production such as developmental language disorders (DLD), Autism Spectrum Disorder (ASD), and Cognitive Communication Disorders (i.e., Traumatic Brain Injury, Aphasia, Dementia). Researchers use LSA to answer questions about the nature of communication disorders and the efficacy of intervention approaches, while clinicians use it as part of the diagnostic process, to profile linguistic strengths and weaknesses for clinical decision making, and to monitor progress in intervention.
Although most researchers and clinicians would agree that analysis of discourse level language is a critical part of a comprehensive assessment for individuals with communication disorders, it is not widely used, particularly in clinical practice. The most frequently reported barrier to the use of LSA is reported to be the time required to elicit, transcribe, code, analyze and interpret the data. These findings are consistent regardless of the age and experience of the respondent or even the country of practice. A number of solutions have been proposed that reduce the time requirements for some of the tasks fundamental to LSA, however, the procedures remain underused by practicing clinicians who treat individuals with communications disorders.
Technological advancements in machine learning and the use of Automatic Speech Recognition offer potential solutions to barriers associated with the use of LSA for children and adults with communication disorders.
This Research Topic welcomes original research papers or high-quality manuscripts covering state-of-the-art methodologies and applications of automated speech recognition and machine learning for eliciting, transcribing, coding, analyzing and/or interpreting discourse-level language data of individuals across the life-span with communication disorders (e.g., DLD, ASD, CCD). Topics of interest include but are not limited to:
- ASR and/or machine learning methods for automatizing elicitation of discourse-samples elicited from children and adults with communication disorders
- ASR and/or machine learning methods for automatizing transcription of discourse-samples elicited from children and adults with communication disorders
- ASR and/or machine learning methods for automatizing coding & analysis of discourse-samples elicited from children and adults with communication disorders
Language sample analysis is a well-supported, evidence-based approach for analyzing discourse-level language abilities of individuals with a variety of communication disorders across the life-span. It has been used to quantify and compare discourse production of neurologically typical individuals to those with impairments that impact language production such as developmental language disorders (DLD), Autism Spectrum Disorder (ASD), and Cognitive Communication Disorders (i.e., Traumatic Brain Injury, Aphasia, Dementia). Researchers use LSA to answer questions about the nature of communication disorders and the efficacy of intervention approaches, while clinicians use it as part of the diagnostic process, to profile linguistic strengths and weaknesses for clinical decision making, and to monitor progress in intervention.
Although most researchers and clinicians would agree that analysis of discourse level language is a critical part of a comprehensive assessment for individuals with communication disorders, it is not widely used, particularly in clinical practice. The most frequently reported barrier to the use of LSA is reported to be the time required to elicit, transcribe, code, analyze and interpret the data. These findings are consistent regardless of the age and experience of the respondent or even the country of practice. A number of solutions have been proposed that reduce the time requirements for some of the tasks fundamental to LSA, however, the procedures remain underused by practicing clinicians who treat individuals with communications disorders.
Technological advancements in machine learning and the use of Automatic Speech Recognition offer potential solutions to barriers associated with the use of LSA for children and adults with communication disorders.
This Research Topic welcomes original research papers or high-quality manuscripts covering state-of-the-art methodologies and applications of automated speech recognition and machine learning for eliciting, transcribing, coding, analyzing and/or interpreting discourse-level language data of individuals across the life-span with communication disorders (e.g., DLD, ASD, CCD). Topics of interest include but are not limited to:
- ASR and/or machine learning methods for automatizing elicitation of discourse-samples elicited from children and adults with communication disorders
- ASR and/or machine learning methods for automatizing transcription of discourse-samples elicited from children and adults with communication disorders
- ASR and/or machine learning methods for automatizing coding & analysis of discourse-samples elicited from children and adults with communication disorders