Selecting an acoustic correlate for automated measurement of American English rhotic production in children

Heather Campbell, Daphna Harel, Elaine Hitchcock, Tara McAllister Byun

Research output: Contribution to journalArticle

Abstract

Purpose: A current need in the field of speech–language pathology is the development of reliable and efficient techniques to evaluate accuracy of speech targets over the course of treatment. As acoustic measurement techniques improve, it should become possible to use automated scoring in lieu of ratings from a trained clinician in some contexts. This study asks which acoustic measures correspond most closely with expert ratings of children’s productions of American English /ɹ/ in an effort to develop an automated scoring algorithm for use in treatment targeting rhotics. Method: A series of ordinal mixed-effects regression models were fit over a large sample of children's productions of words containing /ɹ/ that had previously been rated by three trained clinicians. Akaike/Bayesian Information Criteria were used to select the best-fitting model. Result: Controlling for age, sex, and allophonic contextual differences, the measure that accounted for the most variance in speech rating was F3–F2 distance normalised relative to a sample of age- and sex-matched speakers. Conclusion: We recommend this acoustic measure for use in future automated scoring of children’s production of American English rhotics. We also suggest that computer-based treatment with automated scoring should facilitate increases in treatment dosage by improving options for home practice.

Original languageEnglish (US)
Pages (from-to)1-9
Number of pages9
JournalInternational Journal of Speech-Language Pathology
DOIs
StateAccepted/In press - Aug 8 2017

Fingerprint

Acoustics
Therapeutics
Pathology
American English
Scoring
Rhotics
Rating
Clinicians

Keywords

  • biofeedback therapy
  • Human speech
  • linear-mixed effects models
  • ordinal regression analysis
  • speech pathology
  • speech sound disorders

ASJC Scopus subject areas

  • Research and Theory
  • Otorhinolaryngology
  • Language and Linguistics
  • Speech and Hearing
  • LPN and LVN

Cite this

@article{74f692f90fc6498f96e8a0c3b4148b05,
title = "Selecting an acoustic correlate for automated measurement of American English rhotic production in children",
abstract = "Purpose: A current need in the field of speech–language pathology is the development of reliable and efficient techniques to evaluate accuracy of speech targets over the course of treatment. As acoustic measurement techniques improve, it should become possible to use automated scoring in lieu of ratings from a trained clinician in some contexts. This study asks which acoustic measures correspond most closely with expert ratings of children’s productions of American English /ɹ/ in an effort to develop an automated scoring algorithm for use in treatment targeting rhotics. Method: A series of ordinal mixed-effects regression models were fit over a large sample of children's productions of words containing /ɹ/ that had previously been rated by three trained clinicians. Akaike/Bayesian Information Criteria were used to select the best-fitting model. Result: Controlling for age, sex, and allophonic contextual differences, the measure that accounted for the most variance in speech rating was F3–F2 distance normalised relative to a sample of age- and sex-matched speakers. Conclusion: We recommend this acoustic measure for use in future automated scoring of children’s production of American English rhotics. We also suggest that computer-based treatment with automated scoring should facilitate increases in treatment dosage by improving options for home practice.",
keywords = "biofeedback therapy, Human speech, linear-mixed effects models, ordinal regression analysis, speech pathology, speech sound disorders",
author = "Heather Campbell and Daphna Harel and Elaine Hitchcock and {McAllister Byun}, Tara",
year = "2017",
month = "8",
day = "8",
doi = "10.1080/17549507.2017.1359334",
language = "English (US)",
pages = "1--9",
journal = "International Journal of Speech-Language Pathology",
issn = "1754-9507",
publisher = "Informa Healthcare",

}

TY - JOUR

T1 - Selecting an acoustic correlate for automated measurement of American English rhotic production in children

AU - Campbell, Heather

AU - Harel, Daphna

AU - Hitchcock, Elaine

AU - McAllister Byun, Tara

PY - 2017/8/8

Y1 - 2017/8/8

N2 - Purpose: A current need in the field of speech–language pathology is the development of reliable and efficient techniques to evaluate accuracy of speech targets over the course of treatment. As acoustic measurement techniques improve, it should become possible to use automated scoring in lieu of ratings from a trained clinician in some contexts. This study asks which acoustic measures correspond most closely with expert ratings of children’s productions of American English /ɹ/ in an effort to develop an automated scoring algorithm for use in treatment targeting rhotics. Method: A series of ordinal mixed-effects regression models were fit over a large sample of children's productions of words containing /ɹ/ that had previously been rated by three trained clinicians. Akaike/Bayesian Information Criteria were used to select the best-fitting model. Result: Controlling for age, sex, and allophonic contextual differences, the measure that accounted for the most variance in speech rating was F3–F2 distance normalised relative to a sample of age- and sex-matched speakers. Conclusion: We recommend this acoustic measure for use in future automated scoring of children’s production of American English rhotics. We also suggest that computer-based treatment with automated scoring should facilitate increases in treatment dosage by improving options for home practice.

AB - Purpose: A current need in the field of speech–language pathology is the development of reliable and efficient techniques to evaluate accuracy of speech targets over the course of treatment. As acoustic measurement techniques improve, it should become possible to use automated scoring in lieu of ratings from a trained clinician in some contexts. This study asks which acoustic measures correspond most closely with expert ratings of children’s productions of American English /ɹ/ in an effort to develop an automated scoring algorithm for use in treatment targeting rhotics. Method: A series of ordinal mixed-effects regression models were fit over a large sample of children's productions of words containing /ɹ/ that had previously been rated by three trained clinicians. Akaike/Bayesian Information Criteria were used to select the best-fitting model. Result: Controlling for age, sex, and allophonic contextual differences, the measure that accounted for the most variance in speech rating was F3–F2 distance normalised relative to a sample of age- and sex-matched speakers. Conclusion: We recommend this acoustic measure for use in future automated scoring of children’s production of American English rhotics. We also suggest that computer-based treatment with automated scoring should facilitate increases in treatment dosage by improving options for home practice.

KW - biofeedback therapy

KW - Human speech

KW - linear-mixed effects models

KW - ordinal regression analysis

KW - speech pathology

KW - speech sound disorders

UR - http://www.scopus.com/inward/record.url?scp=85027130312&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027130312&partnerID=8YFLogxK

U2 - 10.1080/17549507.2017.1359334

DO - 10.1080/17549507.2017.1359334

M3 - Article

SP - 1

EP - 9

JO - International Journal of Speech-Language Pathology

JF - International Journal of Speech-Language Pathology

SN - 1754-9507

ER -