Acoustic Variability in the Healthy Female Voice Within and Across Days: How Much and Why?

Normative Values of Client-Reported Outcome Measures and Self-Ratings of Six Voice Parameters via the VoiceEvalU8 App

OBJECTIVES

Ecological momentary assessments (EMA) and interventions (EMI) have the potential to assess and offer interventions repeatedly within the client's daily life through mobile app technology. Considering the development of the EMA VoiceEvalU8, the current work provided normative data by comparing traditional (i.e., paper and pencil) and electronic (i.e., VoiceEvalU8 app) administration modalities of client-reported outcome measures and client self-ratings of six voice parameters twice a day in their functional environment.

METHODS

In Experiment 1, 50 vocally healthy cisgender women and men completed the Voice Handicap Index (VHI)-30, VHI-10, and Vocal Fatigue Index (VFI) with paper and pencil and the VoiceEvalU8 app. The order of administration modality and perceptual scales was randomized for each participant. In Experiment 2, 104 vocally healthy cisgender women and men used the VoiceEvalU8 app across five days in the morning before talking all day and in the evening after talking all day to self-rate six voice parameters (i.e., overall quality, roughness, breathiness, strain, pitch, and loudness) on a scale from 0 - 100. Mann Whitney U tests were run for all measures. Means and standard deviations are reported for descriptive analysis and normative values.

RESULTS

In Experiment 1, results showed no significant differences between paper and pencil and VoiceEValU8 app administration for the VHI-30, VHI-10, and VFI for vocally healthy adults. Normative values in the current study were consistent with previous literature for the VHI-10 and VFI factor 1 and 2, but slightly higher for the VHI-30 and slightly lower for the VFI factor 3. In Experiment 2, results demonstrated significant differences for all six voice parameters between the morning and evening log sessions. Normative values demonstrated higher self-ratings in the morning as compared to the evening.

CONCLUSIONS

For vocally healthy adults, completing the VHI-30, VHI-10, and VFI via paper and pencil and the VoiceEvalU8 app yielded the same results. Client self-ratings of six voice parameters on a scale from 0 - 100 twice a day demonstrated that vocally healthy adults perceived voice to be worse in the morning before talking all day as compared to the evening after talking all day. The results from the current work are promising for EMA via the VoiceEvalU8 app and support the need for continued investigations with clients with voice differences, problems, and/or disorders.

Test-Retest Reliability of the Acoustic Voice Quality Index and the Acoustic Breathiness Index

OBJECTIVE

The acoustic voice quality index (AVQI) and the acoustic breathiness index (ABI) are two validated measurements for the objective-acoustic assessment of voice quality. The aim of this study was to evaluate the relative and absolute reliability of AVQI and ABI as well as detect the natural variability of vocally-healthy voices.

METHODS

Totally, 39 vocally-healthy participants received weekly an AVQI- and ABI measurement using the freeware VOXplot over a period of 3 weeks. This study is a within-subject design to assess significant differences between the test and retest measurements (test and mean of two retests) of AVQI and ABI with the paired sample t test and to investigate the reliability of these two acoustic parameters with the intraclass correlation coefficient (ICC), Bland-Altman plot, and the minimal detectable change (MDC).

RESULTS

No significant differences between the test-retest measurements of AVQI and ABI were revealed (all P-values > 0.05). The relative reliability of AVQI and ABI showed good results (eg, ICC = 0.81, and ICC = 0.92, respectively). Absolute reliability measured with the MDC yielded a test-retest value of 0.78 and 0.66 for the AVQI and ABI, respectively. Deviations within this range may be due to natural variations of the voice (random bias).

CONCLUSION

AVQI and ABI revealed high-reliability results in a test-retest measurement of vocally-healthy participants.

Variability in Voice Characteristics of Female Speakers With Phonotraumatic Vocal Fold Lesions

OBJECTIVE

To examine the variability of lesion characteristics and vocal function in female speakers with phonotraumatic vocal fold lesions (PVFLs).

STUDY DESIGN

Prospective Cohort Study METHODS: Thirty adult female speakers with PVFL who were participating in voice therapy were recruited to complete a multidimensional voice analysis at four time points across 1 month. Analysis comprised self-ratings of effort and vocal function, expert ratings of videostroboscopy and audio recordings, and instrumental evaluation by selected aerodynamic and acoustic parameters. The degree of variability across time for each individual was assessed against a minimal clinically important difference threshold.

RESULTS

A high degree of variability across time was observed for participant self-ratings of perceived effort and vocal function, and for instrumental parameters. The greatest degree of variability was observed in aerodynamic measures of airflow and pressure, and the acoustic parameter semitone range. Comparatively less variability was observed in perceptual evaluation of speech, and lesion characteristics via stroboscopy still images. Findings suggest that individuals with all PVFL types and sizes present with variability in function across time, with the greatest degree of variability in function observed in participants with large lesions and vocal fold polyps.

CONCLUSION

Variability is observed in voice characteristics of female speakers with PVFLs across 1 month despite general stability in lesion presentation, suggesting vocal function can change despite the presence of laryngeal pathology. This study highlights the need to explore individual functional and lesion responses across time to determine potential for change and improvement in both aspects when selecting treatment options.