Frequency Matters: Pitch Accents and Information Status Katrin Schweitzer, Michael Walsh, Bernd M¨ bius, o ¨ Arndt Riester, Antje Schweitzer, Hinrich Schutze University of Stuttgart Stuttgart, Germany .@ims.uni-stuttgart.de Abstract This paper presents the results of a series of experiments which examine the impact of two information status categories (given and new) and frequency of occurrence on pitch accent realisations. More specifically the experiments explore within-type similarity of pitch accent productions and the effect information status and frequency of occurrence have on these productions. The results indicate a significant influence of both pitch accent type and information status category on the degree of withintype variability, in line with exemplartheoretic expectations. The questions directly addressed in this paper are as follows: 1. How are different tokens of a pitch accent type realised? Does frequency of occurrence of the pitch accent type play a role? 2. What effect does information status have on realisations of a pitch accent type? Does frequency of occurrence of the information status category play a role? 3. Does frequency of occurrence in pitch accents and in information status play a role, i.e. is there a combined effect? In examining the realisation of pitch accent tokens, their degree of similarity is the characteristic under investigation. Similarity is calculated by determining the cosine of the angle between pairs of pitch accent vector representations (see section 6). The results in this study are examined from an exemplar-theoretic perspective (see section 3). The expectations within that framework are based upon two different aspects. Firstly, it is expected that, since all exemplars are stored, exemplars of a type that occur often, offer the speaker a wider selection of exemplars to choose from during production (Schweitzer and M¨ bius, 2004), i.e. the o realisations are expected to be more variable than those of a rare type. However, another aspect of Exemplar Theory has to be considered, namely entrenchment (Pierrehumbert, 2001; Bybee, 2006). The central idea here is that frequently occurring behaviours undergo processes of entrenchment, they become in some sense routine. Therefore realisations of a very frequent type are expected to be realised similar to each other. Thus, similarity and variability are expressions of the same characteristic: the higher the degree of similarity of pitch accent tokens, the lower their realisation variability. 1 Introduction It seems both intuitive and likely that prosody should have a significant role to play in marking information status in speech. While there are well established expectations concerning typical associations between categories of information status and categories of pitch accent, e.g. rising LH accents are often a marker for givenness, there is nevertheless some variability here (Baumann, 2006). Furthermore, little research has focused on how pitch accent tokens of the same type are realised nor have the effects of information status and frequency of occurrence been considered. From the perspective of speech technology, the tasks of automatically inferring and assigning information status clearly have significant importance for speech synthesis and speech understanding systems. The research presented in this paper examines a number of questions concerning the relationship between two information status categories (new and given), and how tokens of associated pitch accent types are realised. Furthermore the effect of frequency of occurrence is also examined from an exemplar-theoretic perspective. Proceedings of the 12th Conference of the European Chapter of the ACL, pages 728736, Athens, Greece, 30 March 3 April 2009. c 2009 Association for Computational Linguistics 728 The structure of this paper is as follows: Section 2 briefly examines previous work on the interaction of information status categories and pitch accents. Section 3 provides a short introduction to Exemplar Theory. In this study similarity of pitch accent realisations on syllables, annotated with the information status categories of the words they belong to, is examined using the parametric intonation model (M¨ hler, 1998) which is outlined in o Section 4. Section 5 discusses the corpus employed. Section 6 introduces a general methodology which is used in the experiments in Sections 7, 8 and 9. Section 10 then presents some discussion, conclusions and opportunities for future research. study K¨ gler and F´ ry (2008) found givenness to u e lower the high tones of prenuclear pitch accents and to cancel them out postnuclearly. These findings among others (K¨ gler and F´ ry, 2008) motiu e vate an examination of the acoustic detail of pitch accent shape across different information status categories. The experiments presented here go one step further, however, in that they also investigate potential exemplar-theoretic effects. 3 Exemplar Theory 2 Information Status and Intonation It is commonly assumed that pitch accents are the main correlate of information status1 in speech (Halliday, 1967). Generally, accenting is said to signal novelty while deaccenting signals given information (Brown, 1983), although there is counter evidence: various studies note given information being accented (Yule, 1980; Bard and Aylett, 1999). Terken and Hirschberg (1994) point out that new information can also be deaccented. As for the question of which pitch accent type (in terms of ToBI categories (Silverman et al., 1992)) is typically assigned to different degrees of givenness, Pierrehumbert and Hirschberg (1990) find H to be the standard novelty accent for English, a finding which has also been confirmed by Baumann (2006) and Schweitzer et al. (2008) for German. Given information on the other hand, if accented at all, is found to carry L accent in English (Pierrehumbert and Hirschberg, 1990). Baumann (2006) finds deaccentuation to be the most preferred realisation for givenness in his experimental phonetics studies on German. However, Baumann (2006) points out that H+L has also been found as a marker of givenness in a German corpus study. Previous findings on the corpus used in the present study found LH being the typical marker for givenness (Schweitzer et al., 2008). Leaving the phonological level and examining correlates of information status in acoustic detail, Kohler (1991) reports that in a falling accent, an early peak indicates established facts, while a medial peak is used to mark novelty. In a recent The term information status is used in (Prince, 1992) for the first time. Before that the terms givenness, novelty or information structure were used for these concepts. 1 Exemplar Theory is concerned with the idea that the acquisition of language is significantly facilitated by repeated exposure to concrete language input, and it has successfully accounted for a number of language phenomena, including diachronic language change and frequency of occurrence effects (Bybee, 2006), the emergence of grammatical knowledge (Abbot-Smith and Tomasello, 2006), syllable duration variability (Schweitzer o and M¨ bius, 2004; Walsh et al., 2007), entrenchment and lenition (Pierrehumbert, 2001), among others. Central to Exemplar Theory are the notions of exemplar storage, frequency of occurrence, recency of occurrence, and similarity. There is an increasing body of evidence which indicates that significant storage of language input exemplars, rich in detail, takes place in memory (Johnson, 1997; Croot and Rastle, 2004; Whiteside and Varley, 1998). These stored exemplars are then employed in the categorisation of new input percepts. Similarly, production is facilitated by accessing these stored exemplars. Computational models of the exemplar memory also argue that it is in a constant state of flux with new inputs updating it and old unused exemplars gradually fading away (Pierrehumbert, 2001). Up to now, virtually no exemplar-theoretic research has examined pitch accent prosody (but see Marsi et al. (2003) for memory-based prediction of pitch accents and prosodic boundaries, and Walsh et al. (2008)(discussed below)) and to the authors' knowledge this paper represents the first attempt to examine the relationship between pitch accent prosody and information status from an exemplar-theoretic perspective. Given the considerable weight of evidence for the influence of frequency of occurrence effects in a variety of other linguistic domains it seems reasonable to explore such effects on pitch accent and information sta- 729 tus realisations. For example, what effect might givenness have on a frequently/infrequently occurring pitch accent? Does novelty produce a similar result? The search for possible frequency of occurrence effects takes place with respect to pitch accent shapes captured by the parametric intonation model discussed next. 4 The Parametric Representation of Intonation Events - PaIntE The model approximates stretches of F0 by employing a phonetically motivated model function (M¨ hler, 1998). This function consists of the sum o of two sigmoids (rising and falling) with a fixed time delay which is selected so that the peak does not fall below 96% of the function's range. The resulting function has six parameters which describe the contour and were employed in the analysis: parameters a1 and a2 express the gradient of the accent's rise and fall, parameter b describes the accent's temporal alignment (which has been shown to be crucial in the description of an accent's shape (van Santen and M¨ bius, 2000)), c1 and c2 model o the ranges of the rising and falling amplitude of the accent's contour, respectively, and parameter d expresses the peak height of the accent.2 These six parameters are thus appropriate to describe different pitch accent shapes. For the annotation of intonation the GToBI(S) annotation scheme (Mayer, 1995) was used. In earlier versions of PaIntE, the approximation of the F0 -contour for HL and H was carried out on the accented and postaccented syllables. However, for these accents the beginning of the rise is likely to start at the preaccented syllable. In the current version of PaIntE the window used for the approximation of the F0 -contour for HL and H accents has been extended to the preaccented syllable, so that the parameters are calculated over the span of the accented syllables and its immediate neighbours (unless it is followed by a boundary tone which causes the window to end at the end of the accented syllable). 5 Corpus The experiments that follow (sections 7, 9 and 8), were carried out on German pitch accents from the Further information and illustrations concerning the mechanics of the PaIntE model can be found in M¨ hler and o Conkie (1998). 2 IMS Radio News Corpus (Rapp, 1998). This corpus was automatically segmented and manually labelled according to GToBI(S) (Mayer, 1995). In the corpus, 1233 syllables are associated with an LH accent, 704 with an HL accent and 162 with an H accent. The corpus contains data from three speakers, two female and a male one, but the majority of the data is produced by the male speaker (888 LH accents, 527 HL accents and 152 H accents). In order to maximise the number of tokens, all three speakers were combined. Of the analysed data, 77.92% come from the male speaker. However, it is not necessarily the case that the same percentage of the variability also comes from this speaker: Both, PaIntE and z-scoring (cf. section 6) normalise across speakers, so the contribution from each individual speaker is unclear. The textual transcription of the corpus was annotated with respect to information status using the annotation scheme proposed by Riester (2008). In this taxonomy information status categories reflect the default contexts in which presuppositions are resolved, which include e. g. discourse context, environment context or encyclopaedic context. The annotations are based solely on the written text and follow strict semantic criteria. Given that textual information alone (i.e. without prosodic or speech related information) is not necessarily sufficient to unambiguously determine the information status associated with a particular word, there are therefore cases where words have multiple annotations, reflecting underspecification of information status. However, it is important to note that in all the experiments reported here, only unambiguous cases are considered. The rich annotation scheme employed in the corpus makes establishing inter-annotator agreement a time-consuming task which is currently underway. Nevertheless, the annotation process was set up in a way to ensure a maximal smoothing of uncertainties. Texts were independently labelled by two annotators. Subsequently, a third, more experienced annotator compared the two results and, in the case of discrepancies, took a final decision. In the present study the categories given and new are examined. These categories do not represent a binary distinction but are two extremes from a set of clearly distinguished categories. For the most part they correspond to the categories textually given and brand-new that are used in Bau- 730 mann (2006), but their scope is more tightly constrained. The information status annotations are mapped to the phonetically transcribed speech signals, from which individual syllable tokens bearing information status are derived. Syllables for which one of the PaIntEparameters was identified as an outlier, were removed. Outliers were defined such that the upper 2.5 percentile as well as the lower 2.5 percentile of the data were excluded. This led to a reduced number of pitch accent tokens: 1021 LH accents, 571 HL accents and 134 H accents. Thus, there is a continuum of frequency of occurrence, high to low, from LH to H. With respect to information status, 102 LH accents, 87 HL accents and 21 H accents were unambiguously labelled as new. For givenness the number of tokens is: 114 LH accents, 44 HL accents and 10 H accents. cos(i, j) = i·j i j (2) where i and j are vectors of the same pitch accent category and · represents the dot product. Each comparison between vectors yields a similarity score in the range [-1,1], where -1 represents high dissimilarity and 1 represents high similarity. The experiments that follow examine distributions of token similarity. In order to establish whether distributions differ significantly two different levels of significance were employed, depending on the number of pairwise comparisons performed. When comparing two distributions (i.e. performing one test), the significance level was set to = 0.05. In those cases where multiple tests were carried out (Experiment 1 and Experiment 3), the level of significance was adjusted (Bonferroni correction) according to the following formula: = 1 - (1 - 1 ) n 1 6 General Methodology In the experiments the general methodology for calculation of similarity detailed in this section was employed. For tokens of the pitch accent types LH, HL and H, each token was modelled using the full set of PaIntE parameters. Thus, each token was represented in terms of a 6-dimensional vector. For each of the pitch accent types the following steps were carried out: For each 6-dimensional pitch accent category token calculate the z-score value for each dimension. The z-score value represents the number of standard deviations the value is away from the mean value for that dimension and allows comparison of values from different normal distributions. The z-score is given by: z - scoredim = valuedim - meandim (1) sdevdim (3) where 1 represents the target significance level (set to 0.05) and n represents the number of tests being performed. The Bonferroni correction is often discussed controversially. The main criticism concerns the increased likelihood of type II errors that lead to non-significance of actually significant findings (Pernegger, 1998). Although this conservative adjustment was applied, the statistical tests in this study resulted in significant p-values indicating the robustness of the findings. 7 Experiment 1: Examining frequency of occurrence effects in pitch accents Hence, at this point each pitch accent is represented by a 6-dimensional vector where each dimension value is a z-score. For each token z-scored vector calculate how similar it is to every other z-scored vector within the same pitch accent category, and, in Experiment 2 and 3, with the same information status value (e.g. new), using the cosine of the angle between the vectors. This is given by: In accordance with the general methodology set out in section 6, the PaIntE vectors of pitch accent tokens of types LH, HL, and H were all z-scored and, within each type, every token was compared for similarity against every other token of the same type, using the cosine of the angle between their vectors. In essence, this experiment illustrates how similarly pitch accents of the same type are realised. Figure 1 depicts the results of the analysis. It shows the density plot for each distribution of cosine-similarity comparison values, whereby the 731 Frequency of Occurrence Effects in Pitch Accents 0.8 H*L L*H H* H*L: Frequency of Occurrence Effects in Information Status Categories given new 1.0 Density 0.6 Density 0.4 0.2 0.0 -1.0 -0.5 0.0 0.5 1.0 0.0 -1.0 0.2 0.4 0.6 0.8 -0.5 0.0 0.5 1.0 Cosine-Similarity Comparison Values Cosine-Similarity Comparison Values Figure 1: Density plots for similarity within pitch accent types. All distributions differ significantly from each other. There is a trend towards greater similarity from highfrequency LH to low-frequency H. Figure 2: Density plots for similarity of HL tokens. Tokens of the low-frequency information status category given display greater similarity to each other than those of the highfrequency information status category new. Walsh et al. (2008) also reported significant differences between these distributions, however, there did not appear to be a clear frequency of occurrence effect. The results in the present study differ from their results because the distributions centre around different ranges of the similarity scale clearly indicating that each accent type behaves differently in terms of similarity/variability between the tokens of the respective type. The differences between the two findings can be ascribed to the augmented PaIntE model (section 4). Given the results from this experiment, the next experiment seeks to establish what relationship, if any, exists between information status and pitch accent production variability. distributions can be compared directly irrespective of the different number of data points. An initial observation is that LH tokens tend to be realised fairly variably, the main portion of the distribution is centred around zero. Tokens of HL tend to be produced more similarly (i.e. the distribution is centred around a higher similarity value), and tokens of H more similarly again. These three distributions were tested against each other for significance using the Kolmogorov-Smirnov test ( = 0.017), yielding p-values of p 0.001. Thus there are significant differences between these distributions. What is particularly noteworthy is that a decrease in frequency of occurrence across pitch accent types co-occurs significantly with an increase in within-type token similarity. While the differences between the graphed distributions do not appear to be highly marked the frequency of occurrence effect is nevertheless in keeping with exemplar-theoretic expectations as posited by Bybee (2006) and Schweitzer and M¨ bius (2004), that is, the high frequency of oco currence entails a large number of stored exemplars, giving the speaker the choice from among a large number of production targets. This wider choice leads to a broader range of chosen targets for different productions and thus to more variable realisations of tokens of the same type. 8 Experiment 2: Examining frequency of occurrence effects in information status categories This experiment was carried out in the same manner as Experiment 1 above with the exception that in this experiment a subset of the corpus was employed: only syllables that were unambiguously labelled with either the information status category new or the category given were included in the analyses. The experiment aims to investigate the effect of information status on the similarity/variability of tokens of different pitch accent types. For each pitch accent type, tokens that were labelled with the information status category new 732 L*H: Frequency of Occurrence Effects in Information Status Categories 0.8 given new 0.6 Effect of Information Status Category "new" across Pitch Accent Types H*L L*H H* Density Density 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 -1.0 -0.5 0.0 0.5 1.0 -1.0 -0.5 0.0 0.5 1.0 Cosine-Similarity Comparison Values Cosine-Similarity Comparison Values Figure 3: Density plots for similarity of LH tokens. The curves differ significantly, a trend towards greater similarity is not observable. The number of tokens for both information status categories is comparable. Figure 4: Density plots for similarity of new tokens across three pitch accent types. In comparison to fig. 1 the trend towards greater similarity from high-frequency LH to lowfrequency H is even more pronounced. were compared to tokens labelled as given. Again, a pairwise Kolmogorov-Smirnov test was applied for each comparison ( = 0.05). Figure 2 depicts the results for HL accents. The K-S test yielded a highly significant difference between the two distributions (p 0.001), reflecting the clearly visible difference between the two curves. It is noteworthy here that for HL the information status category new is more frequent than the category given. Indeed, approximately twice as many are labelled as new than those labelled given. Figure 2 illustrates that new HL accents are realised more variably than given ones. That is, again, an increase in frequency of occurrence co-occurs with an increase in similarity, this time at the level of information status. Figure 3 depicts the difference in similarity/variability for LH between new tokens and given tokens. It is clearly visible that the two curves do not differ as much as those under the HL condition. Both curves centre around zero reflecting the fact that for both types the tokens are variable. Although the Kolmogorov-Smirnov test indicates significance ( = 0.05, p = 0.044), the nature of the impact that information status has in this case is unclear. Here again an effect of frequency of occurrence might be the reason for this result. The high frequency of LH accents in general results in a relative high frequency of given LH tokens. So the token number for both types is similar (102 new LH tokens vs. 114 given LH tokens), there is high frequency in both cases, hence variability. These results, particularly in the case of HL (fig. 2) indicate that information status affects pitch accent realisation. The next experiment compares the effect across different pitch accent types. 9 Experiment 3: Examining the effect of information status across pitch accent types This experiment was carried out in the same manner as Experiments 1 and 2 above. For each pitch accent type, figure 4 depicts within-type pitch accent similarity for tokens unambiguously labelled as new. As with Experiments 1 and 2, frequency of occurrence once more appears to play a significant role. Again, all Kolmogorov-Smirnov tests yielded significant results (p < 0.017 in all cases). Indeed, the difference between the distributions of LH, HL, and H similarity plots appears to be considerably more prominent than in Experiment 1 (see fig. 1). This indicates that under the condition of novelty the frequency of occurrence effect is more pronounced. In other words, there is a considerably more noticeable difference across the distributions of LH, HL and H, when nov- 733 Effect of Information Status Category "given" across Pitch Accent Types 1.2 H*L L*H H* tion between the two domains, and for exemplar theory. Returning to the first question posed at the outset in section 1, it is quite clear from Experiment 1 that a certain amount of variability exists when different tokens of the same pitch accent type are produced. It is also clear, from the same experiment, that the frequency of occurrence of the pitch accent type does indeed play a role: with an increase in frequency comes an increase in variability. This result is in line with the exemplartheoretic view that since all exemplars are stored, exemplars of a type that occur often are more variable because they offer the speaker a wider selection of exemplars to choose from during production (Schweitzer and M¨ bius, 2004). Howo ever, with respect to entrenchment (Pierrehumbert, 2001; Bybee, 2006), i.e. the idea that frequently occurring behaviours undergo processes of entrenchment, in Experiment 1 one might expect to see greater similarity in the realisations of LH. However, it is important to note that while tokens of LH are not particularly similar to each other (the bulk of the distribution is around zero (see figure 1)), they are not too dissimilar either. That is, they rest at the midpoint of the similarity continuum produced by cosine calculation, in quite a normal looking distribution. This is not at odds with the idea of entrenchment. As productions of a pitch accent type become more frequent, the distribution of similarity spreads from the right side of the graph (where infrequent and highly similar H tokens lie) leftwards (through HL) to the point where the LH distribution is found. Beyond this point tokens are excessively different. The second question posed in section 1, and addressed in Experiment 2, sought to ascertain the impact, if any, information status has on pitch accent realisation. Distributions of given and new HL similarity scores differed significantly, as did distributions of given and new LH similarity scores, indicating that information status affects realisation. In other words, for both pitch accent types, given and new tokens behave differently. Concerning the frequency of occurrence of the information status categories, certainly in the case of HL the higher frequency new tokens exhibited more variability. In the case of LH similar numbers of new and given tokens, possibly due to the high frequency of LH in general, Density 0.0 -1.0 0.2 0.4 0.6 0.8 1.0 -0.5 0.0 0.5 1.0 Cosine-Similarity Comparison Values Figure 5: Density plots for similarity of given tokens across three pitch accent types. Mid-frequency HL displays greater similarity than high-frequency LH. For lowest frequency H (only 10 tokens) the trend cannot be observed. elty is considered: novelty compounds the frequency of occurrence effect. Figure 5 illustrates results of the same analysis methodology but applied to tokens of pitch accents unambiguously labelled as given. Once again there is a considerable difference between the distributions of LH and HL tokens (p < 0.017). And again, this difference reflects a more pronounced frequency of occurrence effect for given tokens than for all accents pooled (as described in Experiment 1): the information status category given compounds the frequency of occurrence effect for LH and HL. For H the result is not as clear as for the two more frequent accents. The comparison between H and LH results in a significant difference (p < 0.017) whereas the comparison between H and HL is slightly above the conservative significance level (p = 0.0186). Moreover, the distribution is centred between the distributions for LH and HL and it is thus not clear how to interpret this result with respect to a possible frequency of occurrence effect. However, having only ten instances of given H, the explanatory power of these comparisons is questionable. 10 Discussion The experiments discussed above yield a number of interesting results with implications for research in prosody, information status, the interac- 734 Combined Frequency of Occurrence Effect on L*H and H*L given L*H new L*H new H*L given H*L 1.0 -1.0 -0.5 0.0 0.5 1.0 Cosine-Similarity Comparison Values Figure 6: Density plots for similarity of combinations of information status categories given and new with pitch accent types LH and HL. The distributions show a clear trend towards greater similarity form high-frequency "given LH" and "new LH" to mid-frequency "new HL" and low-frequency "given HL". led to visually similar yet significantly different distributions. Once again sensitivity to frequency of occurrence seems to be present, in line with exemplar-theoretic predictions. The final question concerns the possibility of a combined effect of pitch accent frequency of occurrence and information status frequency of occurrence. Figures 4 and 5 depict a clear compounding effect of both information status categories across the different pitch accent types (and their inherent frequencies) when compared to figure 1. Interestingly, the less frequently occurring given appears to have a greater impact, particularly on high frequency LH. Figure 6 displays all possible combinations of LH, HL, given and new. H is omitted in this graph because of the small number of tokens (10 given, 21 new) and the resulting lack of explanatory power. It is evident that an overall frequency of occurrence effect can be observed: "given LH" and "new LH", which have a similar number of instances (114 vs. 102 tokens) both centre around zero and are thus the most leftward skewed curves in the graph. The distribution of "new HL" (87 tokens) shows a trend towards the right hand side of the graph and thus represents greater similarity of the tokens. The distribution of similarity values for the least frequent combination of pitch accent and information status, "given HL" (44 tokens), centres between 0.5 and 1.0 and is thus the most rightward curve in the graph, reflecting the highest similarity between the tokens. These results highlight an intricate relationship between pitch accent production and information status. The information status of the word influences not only the type and shape of the pitch accent (Pierrehumbert and Hirschberg, 1990; Baumann, 2006; K¨ gler and F´ ry, 2008; Schweitzer et u e al., 2008) but also the similarity of tokens within a pitch accent type. Moreover, this effect is well explainable within the framework of Exemplar Theory as it is subject to frequency of occurrence: tokens of rare types are produced more similar to each other than tokens of frequent types. In the context of speech technology, unfortunately the high variability in highly frequent pitch accents has a negative consequence, as the correlation between a certain pitch accent or a certain information status category and the F0 contour is not a one-to-one relationship. However, forewarned is forearmed and perhaps a finer grained contextual analysis might yield more context specific solutions. Density 0.0 0.2 0.4 0.6 0.8 11 Future Work The methodology outlined in section 6 gives a lucid insight into the levels of similarity found in pitch accent realisations. Further insights, however, could be gleaned from a fine-grained examination of the PaIntE parameters. For example, which parameters differ and under what conditions when examining highly variable tokens? Information status evidently plays a role in pitch accent production but the contexts in which this takes place have yet to be examined. In addition, the role of information structure (focus-background, contrast) also needs to be investigated. A further line of research worth pursuing concerns the impact of information status on the temporal structure of spoken utterances and possible compounding with frequency of occurrence effects. References Kirsten Abbot-Smith and Michael Tomasello. 2006. Exemplar-learning and schematization in a usagebased account of syntactic acquisition. The Linguistic Review, 23(3):275290. Ellen G. Bard and M. P. Aylett. 1999. The dissociation of deaccenting, givenness, and syntactic role in 735 spontaneous speech. In Proceedings of ICPhS (San Francisco), volume 3, pages 17531756. Stefan Baumann. 2006. The Intonation of Givenness Evidence from German., volume 508 of Linguistische Arbeiten. Niemeyer, T¨ bingen. Ph.D. thesis, u Saarland University. Gillian Brown. 1983. Prosodic structure and the given/new distinction. In Anne Cutler and D. Robert Ladd, editors, Prosody: Models and Measurements, pages 6777. Springer, New York. Joan Bybee. 2006. From usage to grammar: The mind's response to repetition. Language, 84:529 551. Karen Croot and Kathleen Rastle. 2004. Is there a syllabary containing stored articulatory plans for speech production in English? In Proceedings of the 10th Australian International Conference on Speech Science and Technology (Sydney), pages 376381. Michael A. K. Halliday. 1967. Intonation and Grammar in British English. Mouton, The Hague. Keith Johnson. 1997. Speech perception without speaker normalization: An exemplar model. In K. Johnson and J. W. Mullennix, editors, Talker Variability in Speech Processing, pages 145165. Academic Press, San Diego. Klaus J. Kohler. 1991. Studies in german intonation. AIPUK (Univ. Kiel), 25. Frank K¨ gler and Caroline F´ ry. 2008. Pitch accent u e scaling on given, new and focused constituents in german. Journal of Phonetics. Erwin Marsi, Martin Reynaert, Antal van den Bosch, Walter Daelemans, and V´ ronique Hoste. 2003. e Learning to predict pitch accents and prosodic boundaries in dutch. In Proceedings of the ACL2003 Conference (Sapporo, Japan), pages 489496. J¨ rg Mayer. o 1995. Transcribing German Intonation The Stuttgart System. Technical report, Universit¨ t Stuttgart. http://www.ims.unia stuttgart.de/phonetik/joerg/labman/STGTsystem.html. Gregor M¨ hler and Alistair Conkie. 1998. Parameto ric modeling of intonation using vector quantization. In Third Intern. Workshop on Speech Synth (Jenolan Caves), pages 311316. Gregor M¨ hler. 1998. Describing intonation with a o parametric model. In Proceedings ICSLP, volume 7, pages 28512854. T. V. Pernegger. 1998. What's wrong with Bonferroni adjustment. British Medical Journal, 316:1236 1238. Janet Pierrehumbert and Julia Hirschberg. 1990. The meaning of intonational contours in the interpretation of discourse. In P. R. Cohen, J. Morgan, and M. E. Pollack, editors, Intentions in Communication, pages 271311. MIT Press, Cambridge. Janet Pierrehumbert. 2001. Exemplar dynamics: Word frequency, lenition and contrast. In Joan Bybee and Paul Hopper, editors, Frequency and the Emergence of Linguistic Structure, pages 137157. Amsterdam. Ellen F. Prince. 1992. The ZPG Letter: Subjects, Definiteness and Information Status. In W. C. Mann and S. A. Thompson, editors, Discourse Description: Diverse Linguistic Analyses of a Fund-Raising Text, pages 295325. Amsterdam. Stefan Rapp. 1998. Automatisierte Erstellung von Korpora f¨ r die Prosodieforschung. Ph.D. thesis, IMS, u Universit¨ t Stuttgart. AIMS 4 (1). a Arndt Riester. 2008. A Semantic Explication of Information Status and the Underspecification of the Recipients' Knowledge. In Atle Grønn, editor, Proceedings of Sinn und Bedeutung 12, Oslo. Antje Schweitzer and Bernd M¨ bius. 2004. Exemplaro based production of prosody: Evidence from segment and syllable durations. In Speech Prosody 2004 (Nara, Japan), pages 459462. Katrin Schweitzer, Arndt Riester, Hans Kamp, and Grzegorz Dogil. 2008. Phonological and acoustic specification of information status - a semantic and phonetic analysis. Poster at "Experimental and Theoretical Advances in Prosody", Cornell University. Kim Silverman, Mary Backman, John Pitrelli, Mari Ostendorf, Colin Wightman, Patti Price, Janet Pierrehumbert, and Julia Hirschberg. 1992. Tobi: A standard for Labeling English Prosody. In Proceedings of ICSLP (Banff, Kanada), volume 2, pages 867870, Banff, Canada. Jacques Terken and Julia Hirschberg. 1994. Deaccentuation of words representing `given' information: effects of persistence of grammatical function and surface position. Language and Speech, 37:125 145. Jan P. H. van Santen and Bernd M¨ bius. 2000. A quano titative model of F0 generation and alignment. In A. Botinis, editor, Intonation--Analysis, Modelling and Technology, pages 269288. Kluwer. Michael Walsh, Hinrich Sch¨ tze, Bernd M¨ bius, and u o Antje Schweitzer. 2007. An exemplar-theoretic account of syllable frequency effects. In Proceedings of ICPhS (Saarbr¨ cken), pages 481484. u o Michael Walsh, Katrin Schweitzer, Bernd M¨ bius, and Hinrich Sch¨ tze. 2008. Examining pitch-accent u variability from an exemplar-theoretic perspective. In Proceedings of Interspeech 2008 (Brisbane). Sandra P. Whiteside and Rosemary A. Varley. 1998. Dual-route phonetic encoding: Some acoustic evidence. In Proceedings of ICSLP (Sydney), volume 7, pages 31553158. George Yule. 1980. Intonation and Givenness in Spoken Discourse. Studies in Language, pages 271 286. 736