A Perceptual Method to Rate Dysphonic Voices

Area of Science:

• Speech-language pathology
• Acoustic analysis
• Expert evaluation methods

Background:

• Inter-rater variability in expert evaluations can impact diagnostic accuracy.
• Standardized methods are needed to improve the reliability of speech quality assessments.
• Natural patient vowels (NPVs) present complex acoustic attributes that are challenging to evaluate consistently.

Purpose of the Study:

• To introduce and validate a novel production-matching method for evaluating speech attributes.
• To assess the method's effectiveness in reducing inter-rater variability among experts.
• To evaluate the learning curve and performance of speech-language pathology students using the new method.

Main Methods:

• The proposed method involves adjusting synthetic vowel attributes to match natural patient vowel (NPV) attributes.
• Seven speech-language pathology (SLP) experts evaluated NPVs using both the new method and a standard method.
• Sixty-five SLP students' performance was assessed using the proposed method, comparing their responses to group and expert benchmarks.

Main Results:

• The production-matching method demonstrated reduced dispersion in expert evaluations compared to the standard method.
• A Kappa index was utilized to measure SLP students' performance and consistency.
• Student responses showed alignment with both their peers and experienced expert evaluations.

Conclusions:

• The production-matching method was well-received by both SLP experts and students.
• The method effectively improved expert consensus and reduced variability in evaluations.
• SLP students showed potential for rapid skill acquisition in discriminating complex speech attributes, accelerating professional development.