Automatisierte Sprach- und Sprechanalyse in Forschung und Diagnostik bei Autismus-Spektrum-Störungen: Potenziale und Barrieren

Digitalisierungsprozesse versprechen auch in der Gesundheitsversorgung ein zeit- und kostensparendes Vorgehen. Eine besondere Engstelle stellt der umfassende Diagnostikprozess dar, welcher zur Identifizierung einer Autismus-Spektrum-Störung (ASS) nötig ist. Dieser Beitrag informiert über Potenziale und bestehende Barrieren in der Forschung und diagnostischen Anwendung automatisierter Sprachanalysen zur Identifikation von ASS bei Kindern.

Abstract

As the diagnostic process for identification of autism spectrum disorder (ASD) is time consuming and requires extensive staffing, waiting lists are long. Reliable screening would allow to identify children at risk earlier, reduce the number of necessary full diagnostic and therefore the waiting list, and would ultimately facilitate early intervention. Automated speech and language analyses have been applied in the classification of ASD-associated symptoms in the last decade with encouraging results. This paper outlines the relevant technical principles and presents previous research approaches. An example from a current research project illustrates possible applications for the identification of ASD in children and adolescents through a technical solution.

Schlüsselwörter

Automatische Sprachentwicklungsanalyse - Autismus-Spektrum-Störungen - Automatische Spracherkennung - Diagnostik - Versorgungsforschung

Key Words

Automatic speech analysis - language analysis - autism spectrum disorders - automatic speech recognition - diagnostics - health services research

Publication History

Article published online:
12 June 2023

• Literatur

• 1 Jothi KR, Mamatha VL. A Systematic Review of Machine Learning based Automatic Speech Assessment System to Evaluate Speech Impairment. 3rd International Conference on Intelligent Sustainable Systems (ICISS). 2020 DOI: 10.1109/ICISS49785.2020.9315920
• 2 Chowdhary KR. Automatic Speech Recognition. In: Fundamentals of Artificial Intelligence. New Delhi: Springer; 2020. DOI: 10.1007/978-81-322-3972-7_20
• 3 Pfister B, Kaufmann T. Sprachverarbeitung: Grundlagen und Methoden der Sprachsynthese und Spracherkennung (2. lage. ed.). Berlin, Heidelberg: Springer Vieweg; 2017
• 4 Rumberg L, Gebauer C, Ehlert H. et al. kidsTALC: A Corpus of 3- to 11-year-old German Children’s Connected Natural Speech. INTERSPEECH 2022; 5160-5164 Proceedings 2022 DOI: 10.21437/Interspeech.2022-330.
• 5 Shahin M, Zafar U, Ahmed B. The Automatic Detection of Speech Disorders in Children: Challenges, Opportunities, and Preliminary Results. IEEE J Select Topics Signal Proc 2020; 14: 400-412 DOI: 10.1109/jstsp.2019.2959393.
• 6 Tuschen L. Einsatz von Sprachverarbeitungstechnologien. Sprache Stimme Gehör 2022; 46: 33-39 DOI: 10.1055/a-1706-9078.
• 7 Chowdhary KR. Natural Language Processing. In: Fundamentals of Artificial Intelligence. New Delhi: Springer; 2020. DOI: 10.1007/978-81-322-3972-7_19
• 8 Lena Research Foundation. LENA – Building brains through early talk (2015-2021). www.lena.org (zuletzt abgerufen am 27.01.2023)
• 9 Ganek H, Eriks-Brophy A. Language ENvironment analysis (LENA) system investigation of day long recordings in children: A literature review. Journal of Communication Disorders 2018; 72: 77-85 DOI: 10.1016/j.jcomdis.2017.12.005.
• 10 Kimbrough-Oller D. All-Day Recordings to Investigate Vocabulary Development: A Case Study of a Trilingual Toddler. Communication Disorders Quarterly 2010; 31 : 213-222 DOI: 10.1177/1525740109358628.
• 11 Regelmann E, Scharff Rethfeldt W. Using the Language ENvironment Analysis System to Investigate Lexical Development in a Set of German-Spanish Bilingual Infant Twins. Hochschule Bremen: City University of Applied Sciences; 2017
• 12 Richards JA, Xu D, Gilkerson J. Development and Performance of the LENA Automatic Autism Screen. Boulder, CO: LENA Foundation; 2010
• 13 Carr J, Xu D, Yoshinaga-Itano C. Language ENvironment Analysis Language and Autism Screen and the Child Development Inventory Social Subscale as a Possible Autism Screen for Children Who Are Deaf or Hard of Hearing. Sem Speech Lang 2014; 35,4: 266-275 DOI: 10.1055/s-0034-1389099.
• 14 Olivati AG, assumpção FB, Misquiatt ARN. Acoustic analysis of speech intonation pattern of individuals with Autism Spectrum Disorders. CoDAS 2017; 29 DOI: 10.1590/2317-1782/20172016081.
• 15 https://ideas.care (zuletzt abgerufen am 27.01.2023)