Technological advances in Digital Signal Processing over the last decade have provided applied linguists with a number of computerized applications for speech analysis which can be of benefit to both the researcher and the instructor. This research project explores the techniques of speech spectrography and implements methods of acoustic phonetics to current issues in Second Language Acquisition theory. Specifically, the effects of vowel production in one's native language on the targets in a second language are investigated. Acoustic measurements of English vowels spoken by Japanese students were compared with measurements of native Japanese vowels and American English vowels. In addition, these data were compared with measurements of learner speech from a variety of native language backgrounds. Vowels from both groups of non-native English speakers showed tendencies toward the center of the vowel space. The less-experienced group showed greater token-to-token variability across height parameters than across frontedness parameters while the more experienced group showed no difference for parameters. Both groups exhibited greater frontedness than height variability between speakers which can be explained in part by differences in vocal tract size. In addition, Flege's Speech Learning Model was tested. Data did not support the hypothesis that similar vowels are more difficult to produce than different vowels. ANOVA tests showed that large LI vowel inventories do not advantage learners of languages with many vowels. The results suggest that the unique qualities of L2 speech may have more to do with developmental processes than L 1 interference.
| Identifer | oai:union.ndltd.org:pdx.edu/oai:pdxscholar.library.pdx.edu:open_access_etds-5773 |
| Date | 28 October 1994 |
| Creators | Bailey, Troy D. |
| Publisher | PDXScholar |
| Source Sets | Portland State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Dissertations and Theses |
Page generated in 0.011 seconds