Tyler, M. D. & Cutler, A. (2009). Cross-language differences in cue use for speech segmentation. Journal of the Acoustic Society of America, 126(1), 367–376.
Presentation: Shelly
Summary: Chris
This study aimed to explore the cues that were used in different languages to show prosodic boundaries. The two cues investigated in the present study were final lengthening of vowels and pitch movement. Three- and four-syllable words were used in three conditions to test where and how listeners use these cues. The conditions included (a) transitional probability only condition, (b) left-edge cue condition, and (c) right-edge cue condition. In Experiment 1, 216 subjects who spoke English, French, or Dutch as their native language participated in the study. Vowel final lengthening occurred in the above three conditions. Results showed that all of the subjects benefited from right-edge lengthening regardless of language backgrounds. Experiment 2 had the same population as Experiment 2. The cue manipulated was pitch movement. A peak of 170 Hz was assigned to three conditions. Results showed that speakers of stress-timed languages (English) used the cue at the right edge, whereas speakers of syllable-timed languages (French) used the cue at the left edge. However, Dutch speakers benefited from pitch movement in both positions. It was concluded that final lengthening was a widespread universal cue for all three types of languages. However, pitch cue was weighted differently across languages. It seemed that Dutch speakers heavily relied on pitch cue to detect boundaries. The reason might be that unlike English, which uses both segmental and suprasegmental cues to mark stress, in Dutch only suprasegmental cues were used. As a result, speakers of Dutch became more sensitive to pitch movement than speakers of English.
Presentation: Shelly
Summary: Chris
This study aimed to explore the cues that were used in different languages to show prosodic boundaries. The two cues investigated in the present study were final lengthening of vowels and pitch movement. Three- and four-syllable words were used in three conditions to test where and how listeners use these cues. The conditions included (a) transitional probability only condition, (b) left-edge cue condition, and (c) right-edge cue condition. In Experiment 1, 216 subjects who spoke English, French, or Dutch as their native language participated in the study. Vowel final lengthening occurred in the above three conditions. Results showed that all of the subjects benefited from right-edge lengthening regardless of language backgrounds. Experiment 2 had the same population as Experiment 2. The cue manipulated was pitch movement. A peak of 170 Hz was assigned to three conditions. Results showed that speakers of stress-timed languages (English) used the cue at the right edge, whereas speakers of syllable-timed languages (French) used the cue at the left edge. However, Dutch speakers benefited from pitch movement in both positions. It was concluded that final lengthening was a widespread universal cue for all three types of languages. However, pitch cue was weighted differently across languages. It seemed that Dutch speakers heavily relied on pitch cue to detect boundaries. The reason might be that unlike English, which uses both segmental and suprasegmental cues to mark stress, in Dutch only suprasegmental cues were used. As a result, speakers of Dutch became more sensitive to pitch movement than speakers of English.
No comments:
Post a Comment