Automatisierte Diagnose prosodischer Störungen bei Aphasie mittels künstlicher neuronaler Netze

J. Haring; C. J. Werner; C. Kohlschein; U. D. Peitz; B. Schumann-Werner; J. Niehues

doi:10.14624/NR2202003

Automatisierte Diagnose prosodischer Störungen bei Aphasie mittels künstlicher neuronaler Netze

Translated title of the contribution: Automated diagnosis of prosodic disorders in aphasia using artificial neural networks

J. Haring^*, C. J. Werner, C. Kohlschein, U. D. Peitz, B. Schumann-Werner, J. Niehues

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

The Aachen Aphasia Test (AAT) also records the spontaneous speech performance of a person with aphasia. However, this part can only be evaluated manually by trained personnel. The present work focuses on the automated scoring of one of the six AAT spontaneous speech scales. The possibility of implementing artificial neural networks for the automated identification of abnormalities of the dimension»prosody and articulation« is investigated and different approaches to this are compared. The aim of the study program is to automate the performance of the AAT using computerized methods while maintaining existing quality requirements.

Translated title of the contribution	Automated diagnosis of prosodic disorders in aphasia using artificial neural networks
Original language	German
Pages (from-to)	69-72
Number of pages	4
Journal	Neurologie und Rehabilitation
Volume	28
Issue number	2
DOIs	https://doi.org/10.14624/NR2202003
Publication status	Published - 1 Jan 2022

Keywords

Aachen Aphasia Test
aphasia
artificial neural networks
dysarthria
prosody

Access to Document

10.14624/NR2202003

Cite this

@article{28423417b93b46b296ee81ec85c30df1,

title = "Automatisierte Diagnose prosodischer St{\"o}rungen bei Aphasie mittels k{\"u}nstlicher neuronaler Netze",

abstract = "The Aachen Aphasia Test (AAT) also records the spontaneous speech performance of a person with aphasia. However, this part can only be evaluated manually by trained personnel. The present work focuses on the automated scoring of one of the six AAT spontaneous speech scales. The possibility of implementing artificial neural networks for the automated identification of abnormalities of the dimension»prosody and articulation« is investigated and different approaches to this are compared. The aim of the study program is to automate the performance of the AAT using computerized methods while maintaining existing quality requirements.",

keywords = "Aachen Aphasia Test, aphasia, artificial neural networks, dysarthria, prosody",

author = "J. Haring and Werner, {C. J.} and C. Kohlschein and Peitz, {U. D.} and B. Schumann-Werner and J. Niehues",

note = "Publisher Copyright: | {\textcopyright} Hippocampus Verlag 2022",

year = "2022",

month = jan,

day = "1",

doi = "10.14624/NR2202003",

language = "German",

volume = "28",

pages = "69--72",

journal = "Neurologie und Rehabilitation",

issn = "0947-2177",

publisher = "Hippocampus Verlag",

number = "2",

}

TY - JOUR

T1 - Automatisierte Diagnose prosodischer Störungen bei Aphasie mittels künstlicher neuronaler Netze

AU - Haring, J.

AU - Werner, C. J.

AU - Kohlschein, C.

AU - Peitz, U. D.

AU - Schumann-Werner, B.

AU - Niehues, J.

N1 - Publisher Copyright: | © Hippocampus Verlag 2022

PY - 2022/1/1

Y1 - 2022/1/1

N2 - The Aachen Aphasia Test (AAT) also records the spontaneous speech performance of a person with aphasia. However, this part can only be evaluated manually by trained personnel. The present work focuses on the automated scoring of one of the six AAT spontaneous speech scales. The possibility of implementing artificial neural networks for the automated identification of abnormalities of the dimension»prosody and articulation« is investigated and different approaches to this are compared. The aim of the study program is to automate the performance of the AAT using computerized methods while maintaining existing quality requirements.

AB - The Aachen Aphasia Test (AAT) also records the spontaneous speech performance of a person with aphasia. However, this part can only be evaluated manually by trained personnel. The present work focuses on the automated scoring of one of the six AAT spontaneous speech scales. The possibility of implementing artificial neural networks for the automated identification of abnormalities of the dimension»prosody and articulation« is investigated and different approaches to this are compared. The aim of the study program is to automate the performance of the AAT using computerized methods while maintaining existing quality requirements.

KW - Aachen Aphasia Test

KW - aphasia

KW - artificial neural networks

KW - dysarthria

KW - prosody

UR - http://www.scopus.com/inward/record.url?scp=85131674039&partnerID=8YFLogxK

U2 - 10.14624/NR2202003

DO - 10.14624/NR2202003

M3 - Article

SN - 0947-2177

VL - 28

SP - 69

EP - 72

JO - Neurologie und Rehabilitation

JF - Neurologie und Rehabilitation

IS - 2

ER -

Automatisierte Diagnose prosodischer Störungen bei Aphasie mittels künstlicher neuronaler Netze

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this