Developing a weighted measure of speech sound accuracy

Area of Science:

• Speech-language pathology
• Linguistics
• Child development

Background:

• Accurate phonetic transcription is crucial for assessing speech sound disorders.
• Existing measures may not fully capture nuances in phonetic accuracy.
• Quantifying speech sound errors requires reliable and valid metrics.

Purpose of the Study:

• To develop and validate a numerical system for quantifying phonetic accuracy in children's speech.
• Introduce the Weighted Speech Sound Accuracy (WSSA) score.
• Evaluate the WSSA score's reliability and validity for differentiating speech sound disorders.

Main Methods:

• Analysis of phonetic transcriptions from diverse child speech samples (toddlers to adolescents, typical and disordered).
• Validation of the WSSA score against established measures and listener judgments.
• Assessment of the WSSA score's sensitivity to changes in phonetic accuracy over time.

Main Results:

• WSSA scores demonstrated strong correlations with existing phonetic accuracy measures and perceived disorder severity.
• The WSSA score successfully differentiated between children with and without speech sound disorders.
• The measure showed high reliability across transcribers, word sets, and testing intervals, capturing developmental growth.

Conclusions:

• The Weighted Speech Sound Accuracy (WSSA) score shows preliminary evidence of being a valid and reliable tool.
• The WSSA score is a promising measure for assessing phonetic accuracy in pediatric speech.
• This system offers a nuanced approach to quantifying speech sound production in children.