This paper presents a longitudinal analysis of the extent to which age affects F 0 and formant frequencies. Five speakers at two time intervals showed a clear effect for F 0 and F1 but no systematic effects for F 2 or F 3. In two speakers for which recordings were available in successive years over a 50 year period, results showed with increasing age a decrease in both F 0 and F1 for a female speaker and a V-shaped pattern, i.e. a decrease followed by an increase in both F 0 and F1 for a male speaker. This analysis also provided strong evidence that F1 approximately tracked F 0 across the years: i.e., the rate of change of (the logarithm of) F 0 and F1 were generally the same. We then also tested that the changes in F1 were not an acoustic artifact of changing F 0. Perception experiments with the main aim of assessing whether changes in F1 contributed to age judgments beyond those from F 0 showed that the contribution of F1 was inconsistent and negligible. The general conclusion is that age-related changes in F1 may be compensatory to offset a physiologically induced decline in F 0 and thereby maintain a relatively constant auditory distance between F 0 and F1.
- Longitudinal analysis
- Vocal aging