Prosodic Phrasing and Syllable Prominence in Spoken Prose. A Validated Coding Manual

Current systems for predicting prosodic prominence and boundaries in texts focus on syntax/semantic-based automatic decoding of sentences that need to be annotated syntactically (Atterer & Klein 2002; Windmann et al. 2011). However, to date, there is no phonetically validated replicable system for manually coding prosodic boundaries and syllable prominence in longer sentences or texts. Based on work in the fields of metrical phonology (Liberman & Prince 1977), phrase formation (Hayes 1989) and existing pause coding systems (Gee and Grosjean 1983), we developed a manual for coding prosodic boundaries (with 6 degrees of juncture) and syllable prominence (8 degrees). Three independent annotators applied the coding system to the beginning pages of four German novels and to four short stories (20 058 syllables, Fleiss kappa .82). For the phonetic validation, eight professional speakers read the excerpts of the novels aloud. We annotated the speech signal automatically with MAUS (Schiel 1999). Using PRAAT (Boersma & Weenink 2019), we extracted pitch, duration, and intensity for each syllable, as well as several phonetic parameters for pauses, and compared all measures obtained to the theoretically predicted levels of syllable prominence and prosodic boundary strength. The validation with the speech signal shows that our annotation system reliably predicts syllable prominence and prosodic boundaries. Since our annotation works with plain text, there are many potential applications of the coding system, covering research on prose rhythm, synthetic speech and (psycho)linguistic research on prosody.

The AMAR mechanism: nominal expressions in the Bantu languages are shaped by apposition and reintegration

Nominal expressions in the Bantu languages have extraordinary typological characteristics. Their word order patterns are extremely diverse and some of the attested patterns are crosslinguistically very rare, or even unique. The same diversity can be found in the number of agreement marker paradigms. Equally remarkable are the prosodic idiosyncrasies found at the level of nominal expressions, especially the existence of prosodic boundaries associated with certain types of adnominal modifiers. Although logically unrelated, I argue that these typological characteristics can be accounted for by a single diachronic scenario here called the AMAR mechanism: a double tendency in the Bantu languages for the emergence of construals in which a nominalized modifier is in apposition to the phrase that contains its semantic head and for such appositional construals to be gradually reintegrated into a single nominal constituent. This paper aims to summarize some of the more remarkable typological characteristics of nominal expressions in the Bantu languages and to lay out the AMAR mechanism as a hypothetical diachronic explanation for many of them.

Reading Fluency in Children and Adolescents Who Stutter

Speech fluency is a major challenge for young persons who stutter. Reading aloud, in particular, puts high demands on fluency, not only regarding online text decoding and articulation, but also in terms of prosodic performance. A written text has to be segmented into a number of prosodic phrases with appropriate breaks. The present study examines to what extent reading fluency (decoding ability, articulation rate, and prosodic phrasing) may be altered in children (9–12 years) and adolescents (13–17 years) who stutter compared to matched control participants. Read speech of 52 children and adolescents who do and do not stutter was analyzed. Children and adolescents who stutter did not differ from their matched control groups regarding reading accuracy and articulation rate. However, children who stutter produced shorter pauses than their matched peers. Results on prosodic phrasing showed that children who stutter produced more major phrases than the control group and more intermediate phrases than adolescents who stutter. Participants who stutter also displayed a higher number of breath pauses. Generally, the number of disfluencies during reading was related to slower articulation rates and more prosodic boundaries. Furthermore, we found age-related changes in general measures of reading fluency (decoding ability and articulation rate), as well as the overall strength of prosodic boundaries and number of breath pauses. This study provides evidence for developmental stages in prosodic phrasing as well as for alterations in reading fluency in children who stutter.

Relativized Prosodic Domains: A Late-Insertion Account of German Plurals

In late-insertion, realizational models of morphology such as Distributed Morphology (DM), the insertion of Vocabulary Items (VIs) is conditioned by cyclic operations in the syntax. This paper explores whether an isomorphic relationship can be established between cyclic operations such as phases and prosodic domains. In the spirit of D’Alessandro and Scheer’s (2015) proposal of a Modular Phase Impenetrability Condition (MPIC), we strive to provide an analysis in which prosodic boundaries in even smaller, word-level-like syntactic structures—the ‘lexical domain’—can be identified solely within the syntax. We propose a DM-account for the distribution of nominal plural exponency in German, which reveals a dominant trend for a trochaic-foot structure for all but -s-plural exponents (Wiese 2001, 2009). Inspired by Gouskova’s (2019) and Svenonius’ (2016) work concerning the prosody–morphology interface, we argue that the index of a Prosodic Word ω in non-s-plurals is associated with a specific feature configuration. We propose that only a n[+pl(ural)] configuration, in which the nominalizing head n hosts the SynSem-feature Num(ber)[+pl(ural)], rather than a general cyclic categorizing phase head such as n, indexes a Prosodic Word ω for nominal plural exponents in (Standard) German. Based on this empirical evidence from German plural exponency, we argue that (i) prosodic boundaries can be established directly by syntactic structures, (ii) these prosodic boundaries condition VI insertion during the initial stages of Spell-Out, and (iii) prosodic domains are based on individual languages’ syntactic structures and feature configurations, and are thus relativized and language-specific in nature.

How focus particles and accents affect attachment

This project shows that focus and information structure, as indicated by the focus particle “only” and pitch accents, influence syntactic attachment, in contrast to the well-known effects of prosodic boundaries on attachment. One written questionnaire, one completion study, and several auditory questionnaires show that the position of “only” strongly affects attachment preferences in ambiguous sentences, while contrastive pitch accents have smaller effects. The two types of focus marking do not interact but independently impact attachment. These results support a modified version of the Focus Attraction Hypothesis, with ambiguous material drawn to attach to the most important information in a sentence. This research shows that information structure can affect sentence structure as well as discourse coherence.

Focus Attracts Attachment

There is growing evidence that pitch accents as well as prosodic boundaries can affect syntactic attachment. But is this an effect of their perceptual salience (the Salience Hypothesis), or is it because accents mark the position of focus (the Focus Attraction Hypothesis)? A pair of auditory comprehension experiments shows that focus position, as indicated by preceding wh-questions instead of by pitch accents, affects attachment by drawing the ambiguous phrase to the focus. This supports the Focus Attraction Hypothesis (or a pragmatic version of salience) for both these results and previous results of accents on attachment. These experiments show that information structure, as indicated with prosody or other means, influences sentence interpretation, and suggests a view on which modifiers are drawn to the most important information in a sentence.

Automatic detection of prosodic boundaries in spontaneous speech

Automatic speech recognition (ASR) and natural language processing (NLP) are expected to benefit from an effective, simple, and reliable method to automatically parse conversational speech. The ability to parse conversational speech depends crucially on the ability to identify boundaries between prosodic phrases. This is done naturally by the human ear, yet has proved surprisingly difficult to achieve reliably and simply in an automatic manner. Efforts to date have focused on detecting phrase boundaries using a variety of linguistic and acoustic cues. We propose a method which does not require model training and utilizes two prosodic cues that are based on ASR output. Boundaries are identified using discontinuities in speech rate (pre-boundary lengthening and phrase-initial acceleration) and silent pauses. The resulting phrases preserve syntactic validity, exhibit pitch reset, and compare well with manual tagging of prosodic boundaries. Collectively, our findings support the notion of prosodic phrases that represent coherent patterns across textual and acoustic parameters.

Phonetic variation in the Korean liquid phoneme

Most discussions of the Korean liquid phoneme /l/ identify two allophones: a flap, [ɾ], in the onset of syllables, and an alveolar lateral approximant syllable-finally and in geminates. However, some research paints a more complex picture indicating a wide range of interspeaker variation for the precise articulatory realization of these allophones. The present research finds that in regards to the tap and laterals realizations previous descriptions are largely correct. It also affirms through analysis of F2 values that previous findings showing that the Korean lateral is palatalized before high front vocoids are correct. Most importantly, it analyzes F3 values to show that the retroflex variant is particularly prevalent near pauses, suggesting that retroflexion may be a secondary cue to prosodic boundaries.

Rhythm Comes, Rhythm Goes: Short-Term Periodicity of Prosodic Phrasing

Speech is perceived as a sequence of meaningful units. Speech prosody helps to delimit these units through pauses and acoustic modulations of pitch, amplitude and speech rate. These prosodic boundaries subdivide utterances into prosodic phrases. To be understood, prosodic phrases must obey cognitive and neurobiological constraints on the side of the listener. In particular, the neurobiological substrates of speech processing have been argued to operate periodically—with one electrophysiological processing cycle being devoted to the processing of one segment of the speech stream. We hypothesized that when processing is periodic, prosodic phrases should show periodicity as well. We investigated the periodicity of prosodic phrases in a corpus of radio news that has been manually annotated for full intonational and intermediate phrases by human experts. We find that sequences of 2 to 5 intermediate phrases are periodic at 0.8 to 1.6 Hertz within their superordinate intonation phrase. Across utterances, the exact duration of intermediate phrases fluctuates with the duration of superordinate intonation phrases, pointing to a dependence of prosodic time scales. Our findings provide evidence of short-term periodicity of prosodic phrasing within a highly specific range. While the determinants of periodicity are unknown, the results are compatible with an association between elec- trophysiological processing time scales and the phonological rhythms of language as such. This is a further step towards closing the gaps between the neurobiology of language, psycholinguistics, and linguistic description.

Prosody-syntax interface and prosodic phrasing in brazilian portuguese

This paper focuses on the Brazilian Portuguese (PB) prosodic phrasing and has two main goals: (1) to find a correlation between the prosodic constituents boundaries, as described by the Prosodic Hierarchy Theory (Nespor & Vogel, 2007 [1986], a.o.) and the perception and production of spontaneous and reading speech breaks, and (2) to describe the phonological characteristics and the syntactic ranking of perceived and non perceived edges. The corpus under analysis includes 5 extracts both of spontaneous and reading speech lasting about 2 minutes each. The reading speech (LE) emerged from the spontaneous speech (FE) orthographic transcription which was collected from an interview in an informal environment. In the perception test, 11 referees heard the 10 speaking extracts, without punctuation, and marked the perceived breaks in the orthographic transcription of each of them. Both the 5 speakers and the 11 referees were students at UFRJ, born in Rio de Janeiro, and were between 22 and 38 years old.The results point out that the prosodic breaks are mainly perceived at the intonational phrase (I) boundary, regardless of the speech style (FE: 91%; LE 99%). However, in LE, 64% of the foreseen I boundaries, described by the Prosodic Hierarchy Theory, were perceived as breaks, but in FE, just 37% were perceived. The most usual nuclear contour in both styles is H+L* L% (this being the Portuguese neutral declarative contour), but its occurrence frequency at perceived breaks draws a distinction between LE and FE (67% and 30%, respectively). In FE, contours like L+H* H% and L*+H H% are also produced (34%). In general, descendant nuclei in LE are predominant, as well as the edge tone L; in FE, both the descendant and ascendant nuclei distribution and low or high boundaries are similar. After running a statistic test, the appearance of an L edge, as a predictive for perception, was globally significant. Concerning the syntactic boundary, it was statistically checked and the result points out that breaks are mainly perceived at the matrix phrase limit (LE: 59%; FE: 61%,), showing the endurance of the matrix phrase edge/I boundary mapping. In general, FE has proved to have a bigger variation on the relation of predicted, perceived and produced, as it was expected, which was also confirmed by statistics. Therefore, the results show that the foreseen I phrasing is fairly robust in both styles, once only 13% of the predicted I boundaries have not been produced as so, regarding intonation. Besides, just 1,4% of the predicted phonological phrase (f) boundaries (and produced as Is) were perceived as breaks by the referees. With this study one may conclude that LE and FE share the same prosodic grammar, performed by the same type of phonological/syntactic cues; nevertheless, these are more consistent in LE and have a more disperse way in FE, adding to a greater difficulty at the systematic perception of prosodic boundaries in FE than in LE.