This is a book chapter on AM submitted to Stefanie Shattuck-Hufnagel & Jonathan Barnes (e... more This is a book chapter on AM submitted to Stefanie Shattuck-Hufnagel & Jonathan Barnes (eds) for their volume "Prosodic Theory and Practice" (MIT Press).
In Pierrehumbert (1980) three rising accents are posited for English, H*, L+H * and L*+H, but thi... more In Pierrehumbert (1980) three rising accents are posited for English, H*, L+H * and L*+H, but this three-way contrast has often been disputed. In the present experiment, Minnesotan and Southern Californian speakers read two dialogs which included four levels of emphasis. Emphasis lengthened segmental duration, raised the scaling of all tones and delayed H alignment in L+H*. Independently of emphasis effects, in both dialects L+H * and L*+H were used in different contexts and were phonetically distinct, with L*+H showing later alignment of both tones, lower scaling of L, and higher scaling of H. In addition, there were phonetic differences between the two dialects, with Southern Californian showing later alignment than Minnesotan English. On the other hand, our data indicate that Minnesotan speakers may lack the H * : L+H * contrast: in one of the dialogs, Minnesotan speakers used L+H * irrespective of emphasis, while Southern Californian speakers switched from H * to L+H * with incr...
Cross-linguistically, focus is marked by syntactic, morphological and prosodic means. Languages l... more Cross-linguistically, focus is marked by syntactic, morphological and prosodic means. Languages like Italian or Catalan utilize syntactic changes to mark narrow focus (Ladd 2008 and references therein), while Wolof (Rialland & Robert 2001) and Chickasaw (Gordon 2007) rely on focus particles. Perhaps the most common strategy is prosodic focus marking, which takes many forms, including
The pragmatic interpretation of Greek wh-questions with different intonation was tested, by askin... more The pragmatic interpretation of Greek wh-questions with different intonation was tested, by asking participants to listen to questions and bet on two follow-up sentences offering alternative explanations on the question’s purpose (informationor noninformation-seeking). L*+H L-!H% and L+H* L-L% were used and crossed (L+H* L-!H% and L*+H LL%), giving rise to four experiment versions in a between-participant design. Responses from 190 Greek listeners supported previous analyses according to which L*+H L-!H% and L+H* L-L% lead to a preference for information-seeking vs. noninformation seeking interpretations respectively. Responses were affected by both the pitch accent and boundary tone, with the joint contribution being most evident in the “crossed” tunes (L+H* L-!H% and L*+H L-L%). These results also support the notion that accents and edge tones contribute independently to pragmatic meaning, while the successful application of betting as an experimental paradigm supports the idea th...
A major issue in intonation research is modelling fine-grained variability while capturing signif... more A major issue in intonation research is modelling fine-grained variability while capturing significant generalizations needed to guide typology and abstraction. I argue that this remains an unresolved issue because of the assumed direct and invariant relationship between abstract tonal categories and F0, which is treated as intonation’s only exponent. New findings and modelling from my own research programme together with documented typological diversity inform a revised understanding of the relation between abstract intonational structure and phonetic realization. This body of work shows that tonal events are comparable to segments: they are realized by a number of phonetic dimensions that exhibit within-category variability and cross-category overlap, and are in trading relationships with each other. Recognizing the variable realization of tonal events requires that we relax the invariance criterion, and accept that (a) the relationship between intonation and F0 is not straightfor...
Functional Principal Component Analysis (FPCA) was used to model within-category variability and ... more Functional Principal Component Analysis (FPCA) was used to model within-category variability and cross-category overlap of F0 features associated with three Greek pitch accents (H*, L+H*, H*+L), and thus disentangle categorical differences from gradient variability. FPCA is an analysis of curves, returning the dominant modes of curve variation as functions, called Functional Principal Components (PCs); each input curve receives a coefficient for identified PCs, representing the contribution of each PC to that curve’s shape. The three accents, which have distinct pragmatic meanings, were utterance-final in declaratives and elicited from thirteen Greek speakers. PC1 and PC2 captured 87.7% of the data variance. Statistical modelling of the coefficients revealed presence of multiple cues, including overall shape, curve height, and position of curve peak. Though PCs showed cross-category overlap, together they distinguished the three accents, providing evidence that tonal events are real...
Face-to-face speech data collection has been next to impossible globally due to COVID-19 restrict... more Face-to-face speech data collection has been next to impossible globally due to COVID-19 restrictions. To address this problem, simultaneous recordings of three repetitions of the cardinal vowels were made using a Zoom H6 Handy Recorder with external microphone (henceforth H6) and compared with two alternatives accessible to potential participants at home: the Zoom meeting application (henceforth Zoom) and two lossless mobile phone applications (Awesome Voice Recorder, and Recorder; henceforth Phone). F0 was tracked accurately by all devices; however, for formant analysis (F1, F2, F3) Phone performed better than Zoom, i.e. more similarly to H6. Zoom recordings also exhibited unexpected drops in intensity. The results suggest that lossless format phone recordings present a viable option for at least some phonetic studies.
Early Spanish-English bilinguals and English controls were tested on the production and perceptio... more Early Spanish-English bilinguals and English controls were tested on the production and perception of negative, short-lag, and long-lag Voice Onset Time (VOT), VOT types spanning the Spanish and English phonetic categories: phonologically, negative and short-lag VOT stops are distinct phonemes in Spanish, but realizations of voiced stops in English. Dominance was critical: more English-dominant bilinguals produced more short-lag VOT stops in response to negative VOT stimuli, and were also less accurate than more balanced bilinguals at discriminating negative from short-lag VOT. Bilinguals performed similarly to monolinguals overall, but they produced more negative VOT tokens and shorter short-lag VOT in response to negative VOT. Their productions were also less well correlated with perception and showed more variation between individuals. These results highlight the variable nature of bilingual production and perception, and demonstrate the need to consider language dominance, indiv...
Two calling melodies of Polish were investigated, the routine call, used to call someone for an e... more Two calling melodies of Polish were investigated, the routine call, used to call someone for an everyday reason, and the urgent call, which conveys disapproval of the addressee's actions. A Discourse Completion Task was used to elicit the two melodies from Polish speakers using twelve names from one to four syllables long; there were three names per syllable count, and speakers produced three tokens of each name with each melody. The results, based on eleven speakers, show that the routine calling melody consists of a low F0 stretch followed by a rise-fall-rise; the urgent calling melody, on the other hand, is a simple rise-fall. Systematic differences were found in the scaling and alignment of tonal targets: the routine call showed late alignment of the accentual pitch peak, and in most instances lower scaling of targets. The accented vowel was also affected, being overall louder in the urgent call. Based on the data and comparisons with other Polish melodies, we analyze the ro...
The Journal of the Acoustical Society of America, 2017
A word-spotting experiment was conducted to investigate whether rhythmic consistency and phrase-f... more A word-spotting experiment was conducted to investigate whether rhythmic consistency and phrase-final lengthening facilitate performance in Korean. Listeners had to spot disyllabic and trisyllabic words in nonsense strings organized in phrases with either the same or variable syllable count; phrase-final lengthening was absent, or occurring either in all phrases or only in the phrase immediately preceding the target. The results show that, for disyllabic targets, inconsistent syllable count and lengthening before the target led to fewer errors. For trisyllabic targets, accuracy was at ceiling, but final lengthening in all phrases reduced reaction times. The results imply that both rhythmic consistency (i.e., regular syllable count) and phrase-final lengthening play a role in word-spotting and, by extension, in speech processing in Korean, as in other languages. However, the results also reflect the language specific role of prosodic cues. First, the cues here were used primarily wit...
This is a book chapter on AM submitted to Stefanie Shattuck-Hufnagel & Jonathan Barnes (e... more This is a book chapter on AM submitted to Stefanie Shattuck-Hufnagel & Jonathan Barnes (eds) for their volume "Prosodic Theory and Practice" (MIT Press).
In Pierrehumbert (1980) three rising accents are posited for English, H*, L+H * and L*+H, but thi... more In Pierrehumbert (1980) three rising accents are posited for English, H*, L+H * and L*+H, but this three-way contrast has often been disputed. In the present experiment, Minnesotan and Southern Californian speakers read two dialogs which included four levels of emphasis. Emphasis lengthened segmental duration, raised the scaling of all tones and delayed H alignment in L+H*. Independently of emphasis effects, in both dialects L+H * and L*+H were used in different contexts and were phonetically distinct, with L*+H showing later alignment of both tones, lower scaling of L, and higher scaling of H. In addition, there were phonetic differences between the two dialects, with Southern Californian showing later alignment than Minnesotan English. On the other hand, our data indicate that Minnesotan speakers may lack the H * : L+H * contrast: in one of the dialogs, Minnesotan speakers used L+H * irrespective of emphasis, while Southern Californian speakers switched from H * to L+H * with incr...
Cross-linguistically, focus is marked by syntactic, morphological and prosodic means. Languages l... more Cross-linguistically, focus is marked by syntactic, morphological and prosodic means. Languages like Italian or Catalan utilize syntactic changes to mark narrow focus (Ladd 2008 and references therein), while Wolof (Rialland & Robert 2001) and Chickasaw (Gordon 2007) rely on focus particles. Perhaps the most common strategy is prosodic focus marking, which takes many forms, including
The pragmatic interpretation of Greek wh-questions with different intonation was tested, by askin... more The pragmatic interpretation of Greek wh-questions with different intonation was tested, by asking participants to listen to questions and bet on two follow-up sentences offering alternative explanations on the question’s purpose (informationor noninformation-seeking). L*+H L-!H% and L+H* L-L% were used and crossed (L+H* L-!H% and L*+H LL%), giving rise to four experiment versions in a between-participant design. Responses from 190 Greek listeners supported previous analyses according to which L*+H L-!H% and L+H* L-L% lead to a preference for information-seeking vs. noninformation seeking interpretations respectively. Responses were affected by both the pitch accent and boundary tone, with the joint contribution being most evident in the “crossed” tunes (L+H* L-!H% and L*+H L-L%). These results also support the notion that accents and edge tones contribute independently to pragmatic meaning, while the successful application of betting as an experimental paradigm supports the idea th...
A major issue in intonation research is modelling fine-grained variability while capturing signif... more A major issue in intonation research is modelling fine-grained variability while capturing significant generalizations needed to guide typology and abstraction. I argue that this remains an unresolved issue because of the assumed direct and invariant relationship between abstract tonal categories and F0, which is treated as intonation’s only exponent. New findings and modelling from my own research programme together with documented typological diversity inform a revised understanding of the relation between abstract intonational structure and phonetic realization. This body of work shows that tonal events are comparable to segments: they are realized by a number of phonetic dimensions that exhibit within-category variability and cross-category overlap, and are in trading relationships with each other. Recognizing the variable realization of tonal events requires that we relax the invariance criterion, and accept that (a) the relationship between intonation and F0 is not straightfor...
Functional Principal Component Analysis (FPCA) was used to model within-category variability and ... more Functional Principal Component Analysis (FPCA) was used to model within-category variability and cross-category overlap of F0 features associated with three Greek pitch accents (H*, L+H*, H*+L), and thus disentangle categorical differences from gradient variability. FPCA is an analysis of curves, returning the dominant modes of curve variation as functions, called Functional Principal Components (PCs); each input curve receives a coefficient for identified PCs, representing the contribution of each PC to that curve’s shape. The three accents, which have distinct pragmatic meanings, were utterance-final in declaratives and elicited from thirteen Greek speakers. PC1 and PC2 captured 87.7% of the data variance. Statistical modelling of the coefficients revealed presence of multiple cues, including overall shape, curve height, and position of curve peak. Though PCs showed cross-category overlap, together they distinguished the three accents, providing evidence that tonal events are real...
Face-to-face speech data collection has been next to impossible globally due to COVID-19 restrict... more Face-to-face speech data collection has been next to impossible globally due to COVID-19 restrictions. To address this problem, simultaneous recordings of three repetitions of the cardinal vowels were made using a Zoom H6 Handy Recorder with external microphone (henceforth H6) and compared with two alternatives accessible to potential participants at home: the Zoom meeting application (henceforth Zoom) and two lossless mobile phone applications (Awesome Voice Recorder, and Recorder; henceforth Phone). F0 was tracked accurately by all devices; however, for formant analysis (F1, F2, F3) Phone performed better than Zoom, i.e. more similarly to H6. Zoom recordings also exhibited unexpected drops in intensity. The results suggest that lossless format phone recordings present a viable option for at least some phonetic studies.
Early Spanish-English bilinguals and English controls were tested on the production and perceptio... more Early Spanish-English bilinguals and English controls were tested on the production and perception of negative, short-lag, and long-lag Voice Onset Time (VOT), VOT types spanning the Spanish and English phonetic categories: phonologically, negative and short-lag VOT stops are distinct phonemes in Spanish, but realizations of voiced stops in English. Dominance was critical: more English-dominant bilinguals produced more short-lag VOT stops in response to negative VOT stimuli, and were also less accurate than more balanced bilinguals at discriminating negative from short-lag VOT. Bilinguals performed similarly to monolinguals overall, but they produced more negative VOT tokens and shorter short-lag VOT in response to negative VOT. Their productions were also less well correlated with perception and showed more variation between individuals. These results highlight the variable nature of bilingual production and perception, and demonstrate the need to consider language dominance, indiv...
Two calling melodies of Polish were investigated, the routine call, used to call someone for an e... more Two calling melodies of Polish were investigated, the routine call, used to call someone for an everyday reason, and the urgent call, which conveys disapproval of the addressee's actions. A Discourse Completion Task was used to elicit the two melodies from Polish speakers using twelve names from one to four syllables long; there were three names per syllable count, and speakers produced three tokens of each name with each melody. The results, based on eleven speakers, show that the routine calling melody consists of a low F0 stretch followed by a rise-fall-rise; the urgent calling melody, on the other hand, is a simple rise-fall. Systematic differences were found in the scaling and alignment of tonal targets: the routine call showed late alignment of the accentual pitch peak, and in most instances lower scaling of targets. The accented vowel was also affected, being overall louder in the urgent call. Based on the data and comparisons with other Polish melodies, we analyze the ro...
The Journal of the Acoustical Society of America, 2017
A word-spotting experiment was conducted to investigate whether rhythmic consistency and phrase-f... more A word-spotting experiment was conducted to investigate whether rhythmic consistency and phrase-final lengthening facilitate performance in Korean. Listeners had to spot disyllabic and trisyllabic words in nonsense strings organized in phrases with either the same or variable syllable count; phrase-final lengthening was absent, or occurring either in all phrases or only in the phrase immediately preceding the target. The results show that, for disyllabic targets, inconsistent syllable count and lengthening before the target led to fewer errors. For trisyllabic targets, accuracy was at ceiling, but final lengthening in all phrases reduced reaction times. The results imply that both rhythmic consistency (i.e., regular syllable count) and phrase-final lengthening play a role in word-spotting and, by extension, in speech processing in Korean, as in other languages. However, the results also reflect the language specific role of prosodic cues. First, the cues here were used primarily wit...
Uploads
Papers