Using spectral measures to differentiate Mandarin and Korean sibilant fricatives

Kallay, Jeffrey; Holliday, Jeffrey J.

doi:10.21437/interspeech.2012-40

Cited by 5 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All speech sound contrasts make use of multiple co-varying phonetic dimensions. The sounds examined here, Mandarin sibilants, have been described as differing in spectral mean or center of gravity, kurtosis, spectral peak, and second formant of the following vowel, to name only a few (Kallay & Holliday, 2012;Lee-Kim, 2011). While speakers and listeners make use of many phonetic dimensions in contrast realization, dimensions differ in their relative strength.…”

Section: Individual Differences In Production Cue Weightingmentioning

confidence: 99%

“…Acoustic studies of standardized Mandarin have often characterized the sibilants as exhibiting a three-way distinction in spectral center of gravity (COG; also sometimes referred to as spectral mean or the first spectral moment, M1). [ʂ] typically exhibits the lowest COG and [s] the highest, as COG is negatively correlated with length of the front cavity (Kallay & Holliday, 2012;Lee, 1999;Lee-Kim, 2011). COG has been shown to be influenced by coarticulation with following vowels, such that COG is lower when followed by a rounded vowel (Jeng, 2006;Y.…”

Section: Choice Of Cuesmentioning

confidence: 99%

“…While COG tends to be the most common measure used in studies of Mandarin, the sibilants have also been shown to differ in other spectral moments and spectral peak locations (Kallay & Holliday, 2012;Lee-Kim, 2011). In particular, sibilants can exhibit multiple major spectral peaks in rounded vowel contexts, which cannot be directly captured by COG measurement.…”

Section: Choice Of Cuesmentioning

confidence: 99%

See 2 more Smart Citations

Differential Cue Weighting in Mandarin Sibilant Production

Hauser

2023

Lang Speech

View full text Add to dashboard Cite

Individual talkers vary in their relative use of different cues to signal phonological contrast. Previous work provides limited and conflicting data on whether such variation is modulated by cue trading or individual differences in speech style. This paper examines differential cue weighting patterns in Mandarin sibilants as a test case for these hypotheses. Standardized Mandarin exhibits a three-way place contrast between retroflex, alveopalatal, and alveolar sibilants with individual differences in relative weighting of spectral center of gravity (COG) and the second formant of the following vowel (F2). In results from a speech production task, cue weights of COG and F2 are inversely correlated across speakers, demonstrating a trade-off relationship in cue use. These findings are consistent with a cue trading account of individual differences in contrast signaling.

show abstract

Section: Individual Differences In Production Cue Weightingmentioning

confidence: 99%

Section: Choice Of Cuesmentioning

confidence: 99%

Section: Choice Of Cuesmentioning

confidence: 99%

See 1 more Smart Citation

Differential Cue Weighting in Mandarin Sibilant Production

Hauser

2023

Lang Speech

View full text Add to dashboard Cite

show abstract

Second Language Experience Can Hinder the Discrimination of Nonnative Phonological Contrasts

Holliday

2016

Phonetica

View full text Add to dashboard Cite

Background/Aims: Many studies have shown that experienced second language (L2) learners are more skilled than novice L2 learners at a variety of L2 tasks, including auditory discrimination between members of L2 phonological contrasts. In this paper we argued that while L2 experience is typically beneficial when comparing the effects of more versus less experience, it is not necessarily beneficial when comparing the effects of some experience versus none. Methods: We compared the perceptual assimilation and discrimination of the Korean sibilant fricatives /sh/ and /s*/ by 3 native Mandarin populations: naïve listeners, novice L2 learners (4-6 weeks of experience), and advanced L2 learners (over 2 years of experience). Results: The perceptual assimilation of /sh/ changed as a result of L2 experience, but only in the /a/ context. It is also shown that novice L2 learners were less accurate than the naïve listeners at discriminating between /sh/ and /s*/ but, crucially, only in the /a/ context. Conclusion: The perception of /sh/ by L2 learners may be affected by knowledge of the L2 unavailable to naïve listeners, and some aspects of this knowledge may result in a decline in discrimination accuracy after even a very short period of L2 experience.

show abstract

Cross-language correspondences in the face of change: Phonetic independence versus convergence in two Korean-Mandarin bilingual communities

Schertz

Kang

Han

2017

International Journal of Bilingualism

View full text Add to dashboard Cite

Aims and Objectives/Purpose/Research Questions: We investigate the robustness of cross-language phonetic correspondences in two bilingual communities over time, focusing on whether corresponding sounds (e.g. Mandarin /s/ and Korean /s’/) remain coupled in the face of language change, or whether the categories diverge over time in younger, more proficient bilinguals. Design/Methodology/Approach: We quantify the extent of assimilation versus independence of categories across languages by comparing bilinguals’ production of place of articulation and laryngeal contrasts in Mandarin and Korean sibilants. Distinct language-internal changes were expected on each dimension. Data and analysis: 107 speakers varying in age (aged 19–83), gender, and dialect participated in the study. Acoustic measurements (center of gravity of frication, voice onset time) and statistical analyses were performed on a total of ~11,000 tokens. Findings/Conclusions: The extent of cross-language independence differed on the two dimensions. Corresponding segments across the two languages remained tightly coupled in terms of place of articulation, even in the face of change; on the other hand, a language-internal change in the Korean laryngeal contrast left corresponding Mandarin segments unaffected, resulting in divergence of originally corresponding categories. We also found unpredicted changes on each dimension, and these changes progressed concurrently in the two languages. Originality: The study of correspondences in the context of independent sound change provides a unique perspective from which to evaluate the robustness of cross-language interaction, and the parallel analysis of two separate dimensions in two communities adds to the generalizability of results. Significance/Implications: Most changes occurred concurrently in the two languages, suggesting that similar phonetic categories across languages can remain tightly coupled, even in highly proficient bilinguals where phonetic independence is expected. However, one of the primary expected changes (voice onset time merger in Korean) did not affect corresponding segments in Mandarin, indicating that the extent of cross-language independence in phonetic correspondences may differ even within the same population. We discuss potential reasons for the different results.

show abstract

Using spectral measures to differentiate Mandarin and Korean sibilant fricatives

Cited by 5 publications

References 5 publications

Differential Cue Weighting in Mandarin Sibilant Production

Differential Cue Weighting in Mandarin Sibilant Production

Second Language Experience Can Hinder the Discrimination of Nonnative Phonological Contrasts

Cross-language correspondences in the face of change: Phonetic independence versus convergence in two Korean-Mandarin bilingual communities

Contact Info

Product

Resources

About