Assessment of intelligibility using children's spontaneous speech: methodological aspects

Area of Science:

• Speech-Language Pathology
• Communication Sciences and Disorders
• Developmental Pediatrics

Background:

• Assessing speech intelligibility is crucial for understanding communication disorders.
• Spontaneous speech intelligibility reflects overall speech disorder severity.
• A gap exists in reliable methods for measuring spontaneous speech intelligibility.

Purpose of the Study:

• To validate a novel method for assessing spontaneous speech intelligibility.
• To determine the reliability of this intelligibility assessment method.
• To compare intelligibility scores between children with and without speech-sound disorders.

Main Methods:

• Recorded spontaneous speech from children with speech-sound disorders and typically developing peers.
• Listeners transcribed understandable words from recorded speech samples.
• Intelligibility scores were calculated as the percentage of perceived understood syllables.
• Correlations with Percentage of Consonants Correct (PCC) were analyzed.

Main Results:

• The intelligibility scores demonstrated high validity, correlating with PCC.
• Significant differences in intelligibility were found between the two groups.
• Excellent inter-judge reliability was achieved using the mean score across multiple listeners.

Conclusions:

• The proposed method is recommended for assessing speech intelligibility, particularly when using the mean score from several listeners.
• This method can be valuable for tracking intelligibility changes in clinical settings over time.
• Consistent listener assessment is recommended for longitudinal evaluations.