Variation in vowel space across dialects and generations
Robert A. Fox (Lead Investigator)
Ewa Jacewicz
Joseph Salmons
Research Goals: Broadly speaking, a speaker's "acoustic vowel space" represents the area in the F1 by F2 plane in which one finds the vowels produced by that speaker. This work examines the effects that positional and distributional differences in vowels across three different regional dialects of American English have upon the size of the acoustic vowel space. In addition, as our research has shown significant changes in vowel positions and dynamics across different ages of speakers within a regional dialect, we are also looking at changes in vowels space size as a function of age. The speakers in this study came from distinct dialect regions: Southeastern Wisconsin (Northern dialect), Central Ohio (Midlands dialect) and Western North Carolina (Inland South or Appalachian dialect). Speech materials consisted of the following single words of the /hVd/ structure: heed, hid, head, hey'd, had, heard, who'd, hood, hoed, hawed, hod, hide, howed, hoyd. The words contained 14 vowels and diphthongs of American English which were used for calculation of vowel space areas. The frequencies of F1, F2, and F3 were measured at five equidistant temporal locations corresponding to the 20-35-50-65-80%-point over the course of each vowel's duration. These measurements provided a means of ascertaining each speaker's full "working acoustic space" for his or her vowels. The size of the vowel space was calculated on the basis of the polygons formed from the four corner vowels as well as all 14 vowels (multilaterals). In general, the quadrilateral areas significantly underestimate the working vowel area in every dialect for every age group. In addition, there are significant differences between dialects, and some variation across age groups. Traditional methods of determining vowel space areas are insufficient to capture the significant variation found here.
Publications and work in progress
(visit Publications and Presentations pages for available downloads)
Publications
Fox, R. A., and Jacewicz, E. (2008). "Analysis of total vowel space areas in three regional dialects of American English." Proceedings of Acoustics'08: Joint meeting of the Acoustical Society of America, the European Acoustics Association, and the French Acoustical Society, pp. 495-500. Paris, France.
Jacewicz, E., Fox, R. A., and Salmons, J. (2007) "Vowel space areas across dialects and gender." In: Proceedings of the XVIth International Congress of Phonetic Sciences, edited by J. Trouvain and W.J. Barry, pp. 1465-1468. Saarbrucken, Germany.
Recent presentations
Fox, R.A. and Jacewicz, E. (2010). "Variation in vowel space across dialects and generations." To be presented to the 64th Annual Convention of the Ohio Speech-Language-Hearing Association, 11-13 March, Columbus OH.
Robert A. Fox (Lead Investigator)
Ewa Jacewicz
Joseph Salmons
Research Goals: Broadly speaking, a speaker's "acoustic vowel space" represents the area in the F1 by F2 plane in which one finds the vowels produced by that speaker. This work examines the effects that positional and distributional differences in vowels across three different regional dialects of American English have upon the size of the acoustic vowel space. In addition, as our research has shown significant changes in vowel positions and dynamics across different ages of speakers within a regional dialect, we are also looking at changes in vowels space size as a function of age. The speakers in this study came from distinct dialect regions: Southeastern Wisconsin (Northern dialect), Central Ohio (Midlands dialect) and Western North Carolina (Inland South or Appalachian dialect). Speech materials consisted of the following single words of the /hVd/ structure: heed, hid, head, hey'd, had, heard, who'd, hood, hoed, hawed, hod, hide, howed, hoyd. The words contained 14 vowels and diphthongs of American English which were used for calculation of vowel space areas. The frequencies of F1, F2, and F3 were measured at five equidistant temporal locations corresponding to the 20-35-50-65-80%-point over the course of each vowel's duration. These measurements provided a means of ascertaining each speaker's full "working acoustic space" for his or her vowels. The size of the vowel space was calculated on the basis of the polygons formed from the four corner vowels as well as all 14 vowels (multilaterals). In general, the quadrilateral areas significantly underestimate the working vowel area in every dialect for every age group. In addition, there are significant differences between dialects, and some variation across age groups. Traditional methods of determining vowel space areas are insufficient to capture the significant variation found here.
Publications and work in progress
(visit Publications and Presentations pages for available downloads)
Publications
Fox, R. A., and Jacewicz, E. (2008). "Analysis of total vowel space areas in three regional dialects of American English." Proceedings of Acoustics'08: Joint meeting of the Acoustical Society of America, the European Acoustics Association, and the French Acoustical Society, pp. 495-500. Paris, France.
Jacewicz, E., Fox, R. A., and Salmons, J. (2007) "Vowel space areas across dialects and gender." In: Proceedings of the XVIth International Congress of Phonetic Sciences, edited by J. Trouvain and W.J. Barry, pp. 1465-1468. Saarbrucken, Germany.
Recent presentations
Fox, R.A. and Jacewicz, E. (2010). "Variation in vowel space across dialects and generations." To be presented to the 64th Annual Convention of the Ohio Speech-Language-Hearing Association, 11-13 March, Columbus OH.