Cultural and Linguistic Adaptation of Vocal Fatigue Index for Tamil-Speaking Population (VFI-T).

OBJECTIVE: To adapt the vocal fatigue index (VFI) for the Tamil-speaking population. STUDY DESIGN: Cross-sectional research design METHOD: VFI developed originally in English was translated in the Tamil language, a South Indian Dravidian language. Content validity and reverse translation of VFI in Tamil [VFI-T] was done. The final of VFI-T was administered on 251 subjects with voice concerns/ problems and on 100 individuals with perceptually normal voice for validity and reliability measures. RESULTS: Factor analysis of VFI-T yielded four factors as opposed to three factors in the English version of VFI. The four factors were extracted based on Eigenvalue and variance contributed by each factor namely, I-physical discomfort; II-tiredness of voice and avoidance of voice use; III-increased effort and its effect on voice quality, and IV-Improvement of symptoms with rest. Varimax rotation and factor loading values for individual items identified five items under factor I; seven items under factor II; four items under factor III, and three items were preserved under factor IV. Overall Cronbach's alpha coefficient was 0.91 and alpha values for four factors ranged from 0.75 to 0.85. Item-to-total correlation revealed good internal consistency between the items. The median and interquartile range of VFI-T for cases and controls were viz., I-6.00 (10.00); 0.00 (1.00), factor II-12.00 (11.00); 1.00 (3.00), factor III-7.00 (8.00); 1.00 (2.00), and factor IV-6.00 (6.00); 1.00 (3.00), respectively. Mann-Whitney U test revealed a significant difference between the scores of cases and controls (P < 0.00). Logistic regression suggested sensitivity of 96% and a specificity of 79% for the inventory in the Tamil language. CONCLUSION: VFI-T identified and documented vocal fatigue under four factors in subjects with voice disorders.

Cepstral Measures to Analyze Vocal Fatigue in Individuals With Hyperfunctional Voice Disorder.

OBJECTIVE: Cepstral measures are sensitive to slightest deviance in voice quality. Vocal fatigue is an initial symptom of a voice disorder. This study aimed (1) to assess the correlation of vocal fatigue and cepstral measures and (2) to identify a measure for analyzing voice change due to vocal fatigue in individuals with hyperfunctional voice disorders. STUDY DESIGN: Cross-sectional. METHODS: A total of 100 subjects participated in the study, that is Group I (50 cases) and Group II (50 controls). Vocal Fatigue Index (VFI-T) in Tamil was administered to identify the presence of fatigue. Voice samples (sustained vowel and conversational speech) were recorded using the Computerized Speech Lab (Model 4150). Mic to mouth distance was maintained at 10 cm. Extraction of cepstral peak prominence (CPP) and smoothened cepstral peak prominence (CPPS) for a vowel (v) and connected speech (cs) was done using Praat software (version 6.0.39, Boersma & Weenink, 2018). RESULTS: Independent t test was used to compare the cepstral measures between cases and controls. Individuals with vocal fatigue (cases) showed significantly lower cepstral values and higher VFI scores compared to the controls. Spearman's rho revealed moderate to strong negative correlation between factors of VFI and cepstral measures. On stepwise multiple linear regression, CPPcs was retained as a sensitive measure to analyze vocal fatigue in individuals with hyperfunctional voice disorder. CONCLUSION: Cepstral measures can be applied to analyze vocal fatigue.