Auditory-perceptual evaluation of rough and breathy voices: correspondence between analogical visual and numerical scale.

Purpose: To determine the cut-off values of different degrees of vocal roughness and breathiness in an analogical visual scale, from a numerical scale.

Methods: We selected 150 voices from both genders, with quality ranging from neutrality to intense deviations, and predominance of breathiness or roughness. The auditory-perceptual assessment through two scales: 100-mm analogical visual scales and 4-point numerical scale. Intra and interrater reliability of auditory-perceptual ratings was analyzed using the Intraclass Correlation Coefficient. The cut-off values for graduation of roughness and breathiness voices were determinate using the analysis of the Receiver Operating Characteristic (ROC) curve, which is based on the sensitivity, specificity and efficiency.

Results: The intra and interrater reliability of auditory-perceptual ratings was excellent on both scales. The cut-off values (mm) in the analogical visual scale, from the numeric scale, indicated a roughness distribution: 0-8.5 absence of the parameter; 8.5- 28.5 mild presence; 28.5- 59.5 moderate and 59.5- 100 intense. For breathiness, the distribution was: 0-8.5 absence of the parameter; 8.5-33.5 for mild presence; 33.5-52.5 moderate and 52.5-100 intense.

Conclusion: The cut-off values found for the analogical visual scale conclude that the numerical scale zero, which represents the absence of parameter, corresponds to a small range of scores in the analogical visual scale. Furthermore, the third degree of the numerical scale corresponded to a wide range of the analogical visual scale.