21 |
Development and Evaluation of a Korean Treebank and its Application to NLP
|
|
|
|
In: DTIC (2002)
|
|
BASE
|
|
Show details
|
|
23 |
The phonology and phonetics of word -level prosody and its interaction with phrase -level prosody: A study of Korean in comparison to English
|
|
|
|
In: Dissertations available from ProQuest (2002)
|
|
Abstract:
This thesis investigates the following research questions: (1) Does Korean have a metrical structure? (2) If so, what are its acoustic correlates and how do they compare to English? (3) How does it interact with phrasal prosody? In addressing these issues, I first re-examine the identity of the so-called “long” vowel in Korean, and argue that it is a phonetic duration derived from an underlying accent on surface. The phonological argument is based on a reanalysis of what has been traditionally called “vowel shortening” in verb stems and compounds as “accent shift”. I describe phonetic experiments to verify the proposed phonological analysis, comparing the acoustic properties of the so-called “long” and “short” (i.e. stressed and unstressed) vowels of Korean. To compare the results with a well-known stress system, I describe a parallel experiment on English. I employ the following two experimental methods: (1) The location of the target word is varied in three different prosodic positions. (2) The data are analyzed with two complementary methods: Direct Comparison Method (e.g. ‘per’ of ‘perMIT’ vs. ‘PER’ of ‘PERmit’) and Relative Comparison Method (e.g. ‘per’ of ‘perMIT’ vs. ‘MIT’ of ‘perMIT’). The overall results suggest that both Korean and English adopt longer duration, higher fundamental frequency, and greater intensity for the stressed vowels. However, they differ in the details: (1) Korean has a greater phrase final lengthening effect than English. (2) In Korean, the phrase initial rising tone dominates the effects of stress. (3) Pitch plays a more important role in English than in Korean. Finally, I investigate the phrasal prosody and argue the following: (1) Intensification and focus use different phonetic cues (duration and pitch movement, respectively), but both of them respect metrical structure. (2) Vocative chant reflects the special status of the stressed syllable in duration and pitch. In the conclusion, implications of the proposed theory are discussed on deriving the prosodic hierarchy of Korean and the prosodic typology.
|
|
Keyword:
Linguistics
|
|
URL: https://repository.upenn.edu/dissertations/AAI3043899
|
|
BASE
|
|
Hide details
|
|
24 |
Bracketing Guidelines for Penn Korean TreeBank
|
|
|
|
In: IRCS Technical Reports Series (2001)
|
|
BASE
|
|
Show details
|
|
28 |
The Laryngeal Effect in Korean: Phonology or Phonetics?
|
|
|
|
In: University of Pennsylvania Working Papers in Linguistics (2000)
|
|
BASE
|
|
Show details
|
|
|
|