摘要:
This paper presents an electropalatographic (EPG) and acoustic study of
prosodic boundaries effect on the domain-initial segments in Standard
Chinese.1 Two speech sounds, namely, the voiceless unaspirated alveolar
stop /t/ and the high front vowel /i/, were studied to examine the
domain-initial strengthening in both spatial and temporal dimensions. The
articulatory and acoustic parameters of the speech sounds were compared
in initial positions of five prosodic constituents in Standard Chinese,
namely, a Syllable, a Foot, an Immediate Phrase, an Intonational Phrase,
and an Utterance. The results show that: (1) the production of the
domain-initial consonantal gesture was prosodically encoded. The
linguopalatal contact and the seal duration varied as a function of the
prosodic boundary strength. The linguopalatal contact was dependent on
the seal duration in a nonlinear fashion. Of the acoustic properties of the
domain-initial stop, the total voiceless interval and voicing during closure
were found to be reliable acoustic correlates that mark the hierarchical
structure of the prosody. (2) At the release moment of the domain-initial
stop, no consistent pattern was found to support the domain-initial
strengthening. The linguopalatal contact of the vowel immediately
following the domain-initial consonant did not show a clear trend of
domain-initial strengthening; however, the phonatory features of vowels
were indicative of pitch reset at major prosodic boundaries. These indicate that the domain-initial strengthening is restricted on the segment
immediately following the boundary. In conclusion Standard Chinese
strengthens the phonetic features of the domain-initial segments as a
function of boundary strength, which serves as an important way to mark
prosodic structure in Standard Chinese.