Volume 27 Issue 7 - October 10, 2014 PDF
Longitudinal changes of formant values and vowel space: 0-9 years of age
Li-Mei Chen*, Fan- Yin Cheng, and Wei- Chun Hsu
Dept. of Foreign Languages and Literature, National Cheng Kung University
Font Enlarge
This longitudinal study is designed to observe and analyze the spontaneous vocalic production of two Mandarin-learning children (Child A, a boy; Child B, a girl) from birth to 9 years old. All spontaneous productions and elicitation data were analyzed with Kay CSL for formant measurement. The frequency values of the first and the second formants (F1, F2) of the vowels were measured with reference to the four displays: narrowband spectrogram, broadband spectrogram, FFT, and LPC. Developmental vowel production can help to specify the relationship between anatomical differences and the differences of speech production. Moreover, longitudinal study can best describe this developmental trend.


Vowel Formant Frequencies
Standard deviations of F1 and F2 values across ages are calculated to measure articulatory variability (Figures 1 and 2). The results indicate that F1 values were more stable than F2 values in both children. This appears to suggest that children acquire jaw movement sooner than tongue movement.

Vowel Areas
As illustrated in Figures 3 and 4, expand of vowel areas can be found at the early stage of development, followed by reduce of vowel areas at the later stage (4 years old for the boy subject and 6 years old for the girl subject). The broader vowel space at early stages corresponds to the increased variability of vowel formants which might be due to immature motor control (Smith & Goffman, 1998; Green, Moore, Higashikawa, & Steeve, 2000). After acquiring more mature motor control for vowel production, the decreased variability of vowel formants leads to the reduction of F1-F2 space at later stage.

FIGURE 3. Changes of vowel space from birth to 9 years of age in Child A

FIGURE 4. Changes of vowel space from birth to 9 years of age in Child B

Fundamental Frequency
As seen in Figures 5 and 6, there is no clear decline in fundamental frequencies until 8-9 years of age in the boy subject; meanwhile, there is less obvious decrease in f0 in the girl subject. Gender differences can be better identified in formant frequencies (vocal tract resonator), especially F2, rather than f0 (vocal tract length) (Childers & Wu, 1991).


  1. Childers, D. G. & Wu, K. (1991). Gender recognition from speech. Part II: Fine analysis. The Journal of the Acoustical Society of America, 90, 1841-1856.
  2. Fant, G. (1960). Acoustic Theory of Speech Production: With Calculations Based on X-Ray Studies of Russian Articulations. The Hague: Mouton.
  3. Gollin, E. (1981). Development and plasticity. In E. Gollin (ed.), Developmental Plasticity, 231-252. New York: Academic Press.
  4. Green, J. R., Moore, C. A., Higashikawa, M., & Steeve, R. W. (2000). The physiologic development of speech motor control: Lip and jaw coordination. Journal of Speech, Language and Hearing Research, 43, 239-225.
  5. Perry, T. L., Ohde, R. N., & Ashmead, D. H. (2001). The acoustic bases for gender identification from children’s voices. The Journal of the Acoustical Society of America, 109(6), 2988-2998.
  6. Smith, A., & Goffman, L. (1998). Stability and patterning of speech movement sequences in children and adults. Journal of Speech, Hearing and Language Research, 41, 18-30.
  7. Whiteside, S. P., & Hodgson, C. (2000). Some acoustic characteristics in the voices of 6- to 10-year-old children and adults: A comparative sex and developmental perspective. Logopedics, Phoniatrics, Vocology, 25(3), 122-132.
< Previous
Next >
Copyright National Cheng Kung University