| Summary: | 碩士 === 國立清華大學 === 語言學研究所 === 104 === Speech melody and music melody are typically expressed via pitch. However, how
they influence each other has been an issue in both linguistics and musicology for a
long time. Ladd (2013) proposes two categories of lexical tone manifestation in singing,
one being structural and the other phonetic. This thesis considers phonetic
manifestations in examining the differences in expression of tone in Taiwanese and
Mandarin singing.
The first of the experiments discussed here asked native speakers to sing songs that
contain sets of minimal tone pairs under different environments (with accompaniment
or solo). The stimuli were embedded in a song which consists of two variations of a
melody written by Patrick Wong (Wong & Diehl 2002).
The results report on two aspects of tone: F0 and duration. Schellenberg (2013)
examines the phonetic manifestation of tone in singing in Cantonese and Mandarin,
showing that Cantonese singers tend to include a rising contour when singing words
with rising tones; Mandarin singers, however, do not. There are no significant
differences in duration in the two languages. As for Taiwanese, this thesis finds
speakers tend to include a rising contour when singing a word with rising tone.
Moreover, the durations of long tone group (rising and level tones) are longer than those
of the short tone group (falling tones). On the other hand, speakers do not mark any
tone in Mandarin phonetically in either F0 or duration.
The second of the experiments examines whether listeners can correctly determine
the words produced by singers in the singing experiment. The results suggest that
listeners distinguish the tones in Taiwanese by using the cues which discuss in the first
experiment.
These results are consistent with Schellenberg (2013). The production and
perception studies suggest that the speakers use some strategies to differentiate tones in
Taiwanese but not in Mandarin. This can be the effect of the size of tone inventory
|
|---|
