Objective The evaluation of voice quality is a major component of voice assessment. The aim of the present study was to develop a new multivariate acoustic model for the evaluation of breathiness. Method Concatenated voice samples of continuous speech and the sustained vowel [a:] from 970 subjects with dysphonia and 88 vocally healthy subjects were perceptually judged for breathiness severity. Acoustic analyses were conducted on the same concatenated voice samples after removal of the non-voiced segments of the continuous speech sample. The development of an acoustic model for breathiness was based on stepwise multiple linear regression analysis. Concurrent validity, diagnostic accuracy, and cross validation were statistically verified on the basis of the Spearman rank-order correlation coefficient (rs), several estimates of the receiver operating characteristics plus the likelihood ratio, and iterated internal cross correlations. Results Ratings of breathiness from four experts with moderate reliability were used. Stepwise multiple regression analysis yielded a nine-variable acoustic model for the multiparametric measurement of breathiness (Acoustic Breathiness Index [ABI]). A strong correlation was found between ABI and auditory-perceptual rating (rs = 0.840, P = 0.000). The cross correlations confirmed a comparably high degree of association. Additionally, the receiver operating characteristics and likelihood ratio results showed the best diagnostic outcome at a threshold of ABI = 3.44 with a sensitivity of 82.4% and a specificity of 92.9%. Conclusions This study developed a new acoustic multivariate correlate for the evaluation of breathiness in voice. The ABI model showed valid and robust results and is therefore proposed as a new acoustic index for the evaluation of breathiness.
LINK
Purpose Over the last 5 decades, many acoustic measures have been created to measure roughness and breathiness. The aim of this study is to present a meta-analysis of correlation coefficients (r) between auditory-perceptual judgment of roughness and breathiness and various acoustic measures in both sustained vowels and continuous speech. Method Scientific literature reporting perceptual–acoustic correlations on roughness and breathiness were sought in 28 databases. Weighted average correlation coefficients (r w) were calculated when multiple r-values were available for a specific acoustic marker. An r w ≥ .60 was the threshold for an acoustic measure to be considered acceptable. Results From 103 studies of roughness and 107 studies of breathiness that were investigated, only 33 studies and 34 studies, respectively, met the inclusion criteria of the meta-analysis on sustained vowels. Eighty-six acoustic measures were identified for roughness and 85 acoustic measures for breathiness on sustained vowels, in which 43 and 39 measures, respectively, yielded multiple r-values. Finally, only 14 measures for roughness and 12 measures for breathiness produced r w ≥ .60. On continuous speech, 4 measures for roughness and 21 measures for breathiness were identified, yielding 3 and 6 measures, respectively, with multiple r-values in which only 1 and 2, respectively, had r w ≥ .60. Conclusion This meta-analysis showed that only a few acoustic parameters were determined as the best estimators for roughness and breathiness.
The aim of the present investigation was to evaluate the effect of visual feedback on rating voice quality severity level and the reliability of voice quality judgment by inexperienced listeners. For this purpose two training programs were created, each lasting 2 hours. In total 37 undergraduate speech–language therapy students participated in the study and were divided into a visual plus auditory-perceptual feedback group (V + AF), an auditory-perceptual feedback group (AF), and a control group with no feedback (NF). All listeners completed two rating sessions judging overall severity labeled as grade (G), roughness (R), and breathiness (B). The judged voice samples contained the concatenation of continuous speech and sustained phonation. No significant rater reliability changes were found in the pre- and posttest between the three groups in every GRB-parameter (all p > 0.05). There was a training effect seen in the significant improvement of rater reliability for roughness within the NF and AF groups (all p < 0.05), and for breathiness within the V + AF group (p < 0.01). The rating of the severity level of roughness changed significantly after the training in the AF and V + AF groups (p < 0.01), and the breathiness severity level changed significantly after the training in the V + AF group (p < 0.01). The training of V + AF and AF may only minimally influence the reliability in the judgment of voice quality but showed significant influence on rating the severity level of GRB parameters. Therefore, the use of both visual and auditory anchors while rating as well as longer training sessions may be required to draw a firm conclusion.
LINK
