Figure 2 - uploaded by Bettina Braun
Content may be subject to copyright.
Example realization of an ISQ high-rising contour (upper panel) and an RQ plateau contour (lower panel). The third tier indicates the middle of the stressed vowel and its quality, the fourth layer the boundaries of the accented syllables and the fifth layer the GToBI annotation.
Source publication
This paper reports on the prosody of rhetorical questions (RQs) and information-seeking questions (ISQs) in German for two question types, polar questions and constituent questions (henceforth wh-questions). The results are as follows: Phonologically, polar RQs were mainly realized with H-% (high plateau), while polar ISQs mostly ended in H-^H% (hi...
Context in source publication
Similar publications
Aging in speech production is a multidimensional process. Biological, cognitive, social, and communicative factors can change over time, stay relatively stable, or may even compensate for each other. In this longitudinal work, we focus on stability and change at the laryngeal and supralaryngeal levels in the discourse particle euh produced by 10 ol...
Voice problems are routinely assessed in hospital voice clinics by speech and language therapists (SLTs) who are highly skilled in making audio-perceptual evaluations of voice quality. The evaluations are often presented numerically in the form of five-dimensional ‘GRBAS’ scores. Computerised voice quality assessment may be carried out using digita...
This study examines a set of voice parameters to map objective ranges of voice-source characteristics of healthy male speakers of Common Czech. Objective assessment of voice quality is conducted mainly in speakers with voice pathologies, typically using sustained vowels as basis for measurements. In our study, we focused on non-pathological voices...
Introduction
In this study we intend to use speech analysis to analyze the cognitive impairments caused by pathologies of vascular origin such as diabetes, hypertension, hypercholesterolemia and heart disease, predictors of the development of vascular dementia.
Methods
In this study, 40 participants with mild cognitive impairment were asked to rea...
In face-to-face interactions, the conversational feedback produced by the listener to signal attention and participation to the current speaker is multimodal: in the vocal channel, it consists of verbal expressions (e.g., "yes" or "exactly") and vocalizations without lexical content, such as non-lexical backchannels (e.g., "mhm") and laughter; in t...
Citations
... PolQs in German involve subject-verb inversion, as exemplified in (2). In addition, PolQs typically end in a high-rising (H-^H%) boundary tone (Braun et al., 2019), which is characterized by a steep rise in the upper part of the speaker's vocal range (Grice & Baumann, 2002). A low rising boundary tone (L-H%) is also possible in German PolQs, however, it appears to be less frequent than the highrising one (Braun et al., 2019). ...
... In addition, PolQs typically end in a high-rising (H-^H%) boundary tone (Braun et al., 2019), which is characterized by a steep rise in the upper part of the speaker's vocal range (Grice & Baumann, 2002). A low rising boundary tone (L-H%) is also possible in German PolQs, however, it appears to be less frequent than the highrising one (Braun et al., 2019). The typical NPA in German PolQs is low (L*). ...
In heritage language (HL) research, prosodic aspects of heritage speakers (HSs) speech have only recently started to receive attention and the results have been mixed. While some studies report cross-linguistic influence (CLI) for HL intonation, others consider it robust. The present study contributes to this debate and examines the intonational patterns of polar questions (PolQs) in the two languages of 30 HSs of Italian with German as majority language. This language combination displays overlapping and distinct intonational patterns in PolQs, making intonation potentially subject to CLI in bilingual language acquisition. The results of an elicited production task indicate that HSs maintain a systemic difference between their two languages. In German, HSs produce PolQs in a monolingual like manner. In Italian of HSs, both monolingual-like performance and CLI from German has been observed. Early Italian use is beneficial for monolingual-like outcomes in the HL.
... Introduction P rosody, often referred to as the music of speech, is defined as the organizational structure of speech, including linguistic functions such as tone, intonation, stress, and rhythm (Gussenhoven and Chen, 2020;Ladd, 2008). It has been well-established that prosody plays a key role in sentence processing in both L1 (first or native language) and L2 (second or non-native languages), including lexical activation and segmentation (e.g., Cutler and Butterfield, 1992;Cutler and Norris, 1988;Norris et al., 2006;Sanders et al., 2002), syntactic parsing (e.g., Cole et al., 2010a;Frazier et al., 2006;Hwang and Schafer, 2006;Ip and Cutler, 2022;Lee and Watson, 2011;O'Brien et al., 2014;Roncaglia-Denissen et al., 2014;Schafer et al., 2000), information structure marking (e.g., Birch and Clifton, 1995;Breen et al., 2010;Calhoun, 2010;Clopper and Tonhauser, 2013;Katz and Selkirk, 2011;Kügler and Calhoun, 2020;Namjoshi and Tremblay, 2014;Steedman, 2000;Welby, 2003;Xu, 1999), and pragmatic information signaling such as speech attitudes, acts and emotion (e.g., Braun et al., 2019;Lin et al., 2020;Pell et al., 2011;Prieto, 2015;Repp, 2020). ...
The present study provides a systematic review of prosody research in linguistic journals through a bibliometric analysis. Using the bibliographic data from 2001 to 2021 in key linguistic journals that publish prosody-related research, this study adopted co-citation analysis and keyword analysis to investigate the state of the intellectual structure and the emerging trends of research on prosody in linguistics over the past 21 years. Additionally, this study identified the highly cited authors, articles and journals in the field of prosody. The results offer a better understanding of how research in this area has evolved and where the boundaries of prosody research might be pushed in the future.
... As for the independent variable Speech, we found prosodic differences between the descriptions on the one hand and the confirmation questions on the other. These differences were expected and overall consistent with previous findings on the prosodic characteristics of German sentence mode, see [44] and [45]. We will not describe these characteristics in detail here as they are beyond the scope of our research question. ...
... PQ and WHQ with rising intonation "have special connotations" (such as surprise or impatience) beyond being neutral and purely information seeking ( [4], p. 323, [5], p. 477), while PQs with falling intonation (L%) are neutral. Falling question intonation sets Icelandic apart from, for example, English and German, which typically have H% at least in PQ (e.g., [10], [11], and [12] for German; [13], [14], [15], [16], and [17] for English). A fall to L% is much more frequent in WHQ than in PQ in these languages (e.g., [16], [18] for English). ...
... It shows that Faroese PQs often have a low tonal target between an accentual peak and a high terminus (76%; first two and fourth column in Figure 7). This is unlike the continuous rise from a low (L* H-H%) or high (H* H-H%) accented syllable that we find, for example, in English PQ (e.g., [34]) or German (e.g., [12]), and which has been hypothesised for Faroese by [4]. Figure 8 represents the L+H* L-H% nuclear contour, which in our data is the second most frequent nuclear contour, nicely illustrating the low tonal target between accentual and terminal peak. ...
... Hypothesis 2 concerned PQ and stated, mainly based on [4], p. 325, that Faroese PQ have a low nuclear accent followed by a rise (L* H%; see Figure 2 in Section 1 above). This contour is also the default contour in PQ in related head-prominence intonation languages, among them German ( [12], [36]) and English ( [34]). For Faroese, this hypothesis was not confirmed. ...
... Supported by Biezma and Rawlins (2017), Dehé and Braun (2020) depict three basic characteristics: (i) they do not expect an answer (as mentioned by Han, 2002), (ii) they do not have the feel of an assertion, although they may have "the force of a strong assertion' ' (quirk et al., 1985, p. 825), and (iii) they do not have to be answered, but may be (even if the audience is on the other side of the screen). additionally, while they are uttered, there are interesting changes in the prosody used, which can be observed in this type of questions regardless of the language used (e.g., Dehé & Braun (2020) in English, Braun et al. (2019) in German, and Beyssade & Delais-Roussarie (2022) in French). Regardless of the different languages, they are often used as engagement strategies, as is also the case in other similar research communication presentations (e.g., Hu & Liu, 2018;Carter-Thomas & Rowley-Jolivet, 2020;Hyland & Zou, 2021). ...
Open science, an approach based on making research available and understandable to everyone, is currently attracting considerable attention. Online genres are a well-accepted means of democratizing science and spreading scientific research to reach the widest audience (Luzón & Pérez-Llantada, 2019). This paper explores one of these genres devoted to laypeople: FameLab presentations. These are online 3-minute talks on scientific and/or technological subjects which are part of an international competition. One aim of these talks is to engage the audience, and this strategy can be developed by both different language resources (Hyland & Zou, 2021), and multimodal ones (Fortanet-Gómez & Ruiz-Madrid, 2016; Luzón, 2019). Our study focuses on analyzing how questions are used as an engagement device to attract the audience’s attention, and how they are complemented by multimodal features. Our dataset includes 20 FameLab presentations from the 2020 (10) and 2021 (10) editions, when they became live-stream, pre-recorded events because of the COVID-19 pandemic, unlike the traditional dynamics, when they were delivered as in-person live events. Following prior research (e.g., Thompson, 1998), we identified the questions appearing in our dataset, and found similar results to previous findings in comparable genres. We then conducted a multimodal analysis to determine common features among speakers. The results show the need to consider certain non-verbal features which accompany questions, supporting and emphasizing their engagement function. Our research may help understand how multimodal discursive practices are used to explain science, and how they can be transferred to the classroom of Languages for Specific Purposes.
... In the English language (telephone speech corpus), Banuazizi & Creswell (1999) observed that polar RQs were final falling in 44.1% of cases, whereas ISQs were mainly final rising (89.7%). In German (Wochner et al., 2015), Icelandic (Dehè et al., 2018) and English (Braun et al., 2019) the presence of a falling contour (L%) was dominant in both ISQs and RQs regardless of their syntactic structure. This result does not allow to distinguish between the two types of questions, although different nuclear pitch accents, typically a rise-fall pattern, helped to differentiate them. ...
This research focuses on the intonation of rhetorical questions (RQs) in Italian. RQs are non-prototypical questions characterised by specific pragmatic features such as indirect and obvious meaning and the absence of an informative answer. The basic idea is that prosody, in its various verbal and paraverbal components, provides listeners with cues of illocutionary interpretation of RQs, facilitating their recognition in spontaneous speech. Although in the past the presence of a final falling intonation contour was assumed as a categorical feature of RQs, recent studies showed a more heterogeneous situation which did not exclude the possibility of rising intonation contours. Most of these studies were conducted on elicited speech samples (DCT or Role Play) by directly comparing string-identical information-seeking questions with rhetorical questions. The results obtained in several languages show that the differences between the two question types are never clear-cut and do not equally involve the same prosodic parameters (intonation contour, pitch range, duration). This highlights the presence of language-specific components. However, it is possible that the final results might have been influenced by both the mode of acquisition of the speech samples (which were almost always collected in a controlled environment), and by a certain neutralisation of discursive functions. The present research analyses a corpus of 100 spontaneous oral wh-RQs extracted from radio broadcasts. The study presents a phonological analysis of the intonation of RQs, looking at nuclear pitch accents and boundary tones. The results are then compared with those previously obtained for the same geographical area. L’intonazione delle domande retoriche in italiano Questa ricerca si concentra sull’intonazione delle domande retoriche (RQ) in italiano. Si tratta di domande non prototipiche caratterizzate da specifiche caratteristiche pragmatiche, tra cui la presenza di un significato indiretto e ovvio e l’assenza di una risposta informativa. L’idea di base è che la prosodia, nelle sue varie componenti verbali e paraverbali, fornisca agli ascoltatori degli indizi di interpretazione illocutiva delle RQ, facilitandone il riconoscimento nel parlato spontaneo. Sebbene in passato la presenza di un contorno intonativo discendente finale sia stata assunto come una caratteristica categorica delle RQ, studi recenti hanno mostrato una situazione più eterogenea che non esclude la possibilità di contorni intonativi ascendenti. La maggior parte di questi studi è stata condotta su campioni di parlato elicitato (DCT o Role Play), confrontando direttamente coppie frasali identiche formate da domande sincere e domande retoriche. I risultati ottenuti in diverse lingue mostrano che le differenze tra i due tipi di domanda non sono mai nette e non coinvolgono in egual misura i medesimi parametri prosodici (contorno intonativo, pitch range e durata), evidenziando la presenza di componenti linguo-specifiche. Tuttavia, è possibile che i risultati finali siano stati influenzati dalla modalità di acquisizione dei campioni di parlato (quasi sempre raccolti in un ambiente controllato) e da una certa neutralizzazione delle funzioni discorsive. La presente ricerca analizza un corpus di parlato spontaneo formato da 100 domande retoriche con struttura wh- estratte da trasmissioni radiofoniche. Lo studio fornisce un’analisi fonologica dell’intonazione delle domande, focalizzandosi sugli accenti intonativi nucleari e sui toni di confine. I risultati sono stati successivamente comparati con quelli precedentemente ottenuti per la stessa area geografica.
... Como já referido, as perguntas são enunciados presentes de forma considerável na Fala Dirigida à Criança, por volta de 40% (NEWPORT,1977). Dessa forma, nosso estudo focaliza as interrogativas produzidas na FDC no PB e busca analisá-las a partir das classificações das intenções comunicativas apresentadas por Silva e Santos (2015) e Braun et al. (2018), já explicitadas anteriormente, adaptando-as para o contexto de interação adulto-bebê (cf. NAME; SOSA, 2020). ...
RESUMO: Em muitas sociedades, adultos e crianças mais velhas modificam a fala quando se dirigem a bebês, ao conjunto de características semelhantes encontradas nesse tipo de fala dá-se o nome de Fala Dirigida à Criança (FDC). Entre os enunciados produzidos na FDC, as perguntas aparecem de forma relevante. Neste artigo, buscamos investigar, a partir de um estudo longitudinal, qual seria a intenção comunicativa veiculada através do alto índice de interrogativas direcionadas a crianças em fase inicial da aquisição do Português Brasileiro. Para isso, foram analisadas gravações de interações entre os pais e o bebê, Bibo, distribuídas entre seus 4 e 12 meses de vida. Dos 462 enunciados presentes nos 71 áudios analisados, foram encontradas 174 interrogativas, as quais foram classificadas de acordo com a intenção comunicativa: retóricas, semirretóricas e plenas. A análise dos dados revelou que as perguntas corresponderam a 37,6% dos enunciados. Quanto à incidência de cada tipo de pergunta ao longo do período, as retóricas predominaram entre os 4 e os 6 meses de vida do bebê (72%), dando lugar paulatinamente às semirretóricas, que corresponderam a 80% do total de perguntas aos 7-8 meses; nos meses finais do primeiro ano de vida, as perguntas plenas foram mais presentes e, junto com as semirretóricas, totalizaram 91% aos 12 meses. Os resultados apontam o uso das perguntas pelos adultos, inicialmente, para engajar o bebê na cena comunicativa, passando a ter função de verificar seu entendimento sobre situações cotidianas e de obter informações explícitas, conforme o bebê cresce.
... By far the strongest acoustic cue correlating with the perception of surprise was the longer duration of the utterance. Longer duration was also identified as the main phonetic correlate of French SQs [3] and has been found to correlate with non-canonical questions more generally, in particular rhetorical questions [10,11] and exclamatory questions [12,13,14,15]. The next strongest correlates of perceived surprise were a wider F0 range and lower mean F0. ...
The study examines the perception of surprise questions as compared to information-seeking questions. An online perception experiment was carried out in the PCIbex environment using natural stimuli from 6 different speakers from an earlier production experiment. The task of the participants was to evaluate whether the speaker intended to ask for information or to express surprise. The overall correct identification score for information-seeking questions was 84% and for surprise questions 55%, but there was large variability between speakers and items. Higher perception rate of surprise questions is above all related with a longer duration, a wider pitch range, and a lower mean pitch of the utterance.
... 151-160). However, at least for European languages, there is variability among L1 speakers, with no one-to-one mapping between tune and question type, reflecting, e.g., epistemic stance and politeness (Bartels, 1999;Braun et al., 2019;Frota & Prieto, 2015). Production studies show these patterns are difficult to acquire, reflecting L1 transfer on the LILt semantic and systemic dimensions (e.g., Gabriel & Kireva, 2014;Kainada & Lengeris, 2015;McGory, 1997). ...
A great majority of people around the world know more than one language. So, how does knowing one language affect the learning and use of additional languages? The question of cross-language influences is the focus of this book. Do bilinguals hear, understand, and produce language and meaning differently because of the languages they speak? How well can theoretical and computational models of language processing and acquisition explain and predict bilingual use patterns and acquisition trajectories? What learner, language, and context characteristics influence bilingual comprehension and production? This book provides a state-of-the-art review and critique of research into cross-language influences in phonology, lexicon, and morphosyntax, and suggests directions for future research. The interdisciplinary nature of the book bridges the gap between research on bilingualism and second language acquisition. The book will be of interest to graduate students, teachers, and researchers in linguistics and second language acquisition, cognitive psychology, and language education.
... Indeed, L-L% contours, often associated to assertion, seem to be more frequent in rhetorical questions (Bartels, 1999;Han, 2002) while questions ending with an H-H% contour can have both readings. For German, Braun et al. (2019) showed that rhetorical questions presented a high proportion of high plateaus (H-%), a longer duration, and a breathier voice than information-seeking ones. ...
... This goes along with the idea presented by Auchlin and Simon (2004), for whom surprise is part of a positive activation, which contributes to an f0 rising and an augmentation of the tension of the voice. Finally, other phonetic parameters such as voice quality can also contribute to surprise such as breathy voice-see Shochi et al. (2009) for Japanese and Braun et al. (2019) for German-but this has not clearly been shown for French. ...
... Note that sentence lengthening is a common property of both surprise questions and rhetorical ones (cf. Delais-Roussarie and Beyssade, 2019 and Braun et al., 2019). Celle and Pélissier (2020) then conclude that this property characterizes both expressivity and rhetoricity. ...
This paper discusses some semantic/pragmatic aspects of French questions starting with comment 'how' and having an interpretation of reason (similar to reason 'why'), and investigates if and how these properties are reflected by prosody. With reason-comment questions, the speaker doubts that the proposition characterizing the event described in the question can be true, and asks the hearer to help her to revise her expectations about it, by giving her some reasons that make the existence of the event conceivable. The speaker's doubt is often accompanied by additional emotions, such as indignation, exasperation, etc. In order to analyse the prosody of these questions, we conducted a production experiment comparing the reason and the manner reading of comment-questions. Results confirmed that prosodic cues allow distinguishing the two interpretations. In utterances with reason reading, the expression of questioning force is important, though accompanied by expressions of expectation disconfirmation and surprise.
