Publications

Abstract

Most bioacoustics software is used to analyse the already collected acoustics data in batch, i.e., after the data-collecting phase of a scientific study. However, experiments based on animal training require immediate and precise reactions from the experimenter, and thus do not easily dovetail with a typical bioacoustics workflow. Bridging this methodological gap, we have developed a custom application to live-monitor the vocal development of harbour seals in a behavioural experiment. In each trial, the application records and automatically detects an animal's call, and immediately measures duration and acoustic measures such as intensity, fundamental frequency, or formant frequencies. It then displays a spectrogram of the recording and the acoustic measurements, allowing the experimenter to instantly evaluate whether or not to reinforce the animal's vocalisation. From a technical perspective, the rapid and easy development of this custom software was made possible by combining multiple open-source software projects. Here, we integrated the acoustic analyses from Parselmouth, a Python library for Praat, together with PyAudio and Matplotlib's recording and plotting functionality, into a custom graphical user interface created with PyQt. This flexible recombination of different open-source Python libraries allows the whole program to be written in a mere couple of hundred lines of code

Abstract

Listening to a speaker while hearing another speaker talks is a challenging task for elderly listeners. We show that elderly listeners over the age of 65 with various degrees of age-related hearing loss benefit in this situation from also seeing the speaker they intend to listen to. In a phoneme monitoring task, listeners monitored the speech of a target speaker for either the phoneme /p/ or /k/ while simultaneously hearing a competing speaker. Critically, on some trials, the target speaker was also visible. Elderly listeners benefited in their response times and accuracy levels from seeing the target speaker when monitoring for the less visible /k/, but more so when monitoring for the highly visible /p/. Visual speech therefore aids elderly listeners not only by providing segmental information about the target phoneme, but also by providing more global information that allows for better performance in this adverse listening situation.

Abstract

Mutations of the epidermal growth factor receptor (EGFR) gene are found at a relatively high frequency in glioma, with the most common being the de2-7 EGFR (or EGFRvIII). This mutation arises from an in-frame deletion of exons 2-7, which removes 267 amino acids from the extracellular domain of the receptor. Despite being unable to bind ligand, the de2-7 EGFR is constitutively active at a low level. Transfection of human glioma cells with the de2-7 EGFR has little effect in vitro, but when grown as tumor xenografts this mutated receptor imparts a dramatic growth advantage. We mapped the phosphorylation pattern of de2-7 EGFR, both in vivo and in vitro, using a panel of antibodies specific for different phosphorylated tyrosine residues. Phosphorylation of de2-7 EGFR was detected constitutively at all tyrosine sites surveyed in vitro and in vivo, including tyrosine 845, a known target in the wild-type EGFR for src kinase. There was a substantial upregulation of phosphorylation at every yrosine residue of the de2-7 EGFR when cells were grown in vivo compared to the receptor isolated from cells cultured in vitro. Upregulation of phosphorylation at tyrosine 845 could be stimulated in vitro by the addition of specific components of the ECM via an integrindependent mechanism. These observations may partially explain why the growth enhancement mediated by de2-7 EGFR is largely restricted to the in vivo environment

Abstract

Sign language is a rich form of communication, uniquely conveying meaning through a combination of gestures, facial expressions, and body movements. Existing research in sign language generation has predominantly focused on text-to-sign pose generation, while speech-to-sign pose generation remains relatively underexplored. Speech-to-sign language generation models can facilitate effective communication between the deaf and hearing communities. In this paper, we propose an architecture that utilises prosodic information from speech audio and semantic context from text to generate sign pose sequences. In our approach, we adopt a multi-tasking strategy that involves an additional task of predicting Facial Action Units (FAUs). FAUs capture the intricate facial muscle movements that play a crucial role in conveying specific facial expressions during sign language generation. We train our models on an existing Indian Sign language dataset that contains sign language videos with audio and text translations. To evaluate our models, we report Dynamic Time Warping (DTW) and Probability of Correct Keypoints (PCK) scores. We find that combining prosody and text as input, along with incorporating facial action unit prediction as an additional task, outperforms previous models in both DTW and PCK scores. We also discuss the challenges and limitations of speech-to-sign pose generation models to encourage future research in this domain. We release our models, results and code to foster reproducibility and encourage future research1.

Abstract

There is a strong relation between children’s exposure to
spatial terms and their later memory accuracy. In the current
study, we tested whether the production of spatial terms by
children themselves predicts memory accuracy and whether
and how language modality of these encodings modulates
memory accuracy differently. Hearing child speakers of
Turkish and deaf child signers of Turkish Sign Language
described pictures of objects in various spatial relations to each
other and later tested for their memory accuracy of these
pictures in a surprise memory task. We found that having
described the spatial relation between the objects predicted
better memory accuracy. However, the modality of these
descriptions in sign, speech, or speech-plus-gesture did not
reveal differences in memory accuracy. We discuss the
implications of these findings for the relation between spatial
language, memory, and the modality of encoding.

Abstract

This dissertation investigates two seemingly contradictory properties of the speech perception system. On the one hand, listeners are extremely sensitive to the fine phonetic details in the speech signal. These subtle acoustic cues can reduce the temporal ambiguity between words that show initial segmental overlap, and can guide lexical activation. On the other hand, comprehension does not seem to be hampered at all by the drastic reductions that typically occur in casual speech. Complete segments, and sometimes even complete syllables, may be missing, but comprehension is seemingly unaffected. This thesis aims at elucidating how words are represented and accessed in the mental lexicon, by investigating these contradictory phenomena for the domain of morphology

Abstract

Spatial terms in the world’s languages appear to reflect both universal conceptual tendencies and linguistic convention. A similarly mixed picture in the case of color naming has been accounted for in terms of near-optimal partitions of color space. Here, we demonstrate that this account generalizes to spatial terms. We show that the spatial terms of 9 diverse languages near-optimally partition a similarity space of spatial meanings, just as color terms near-optimally partition color space. This account accommodates both universal tendencies and cross-language differences in spatial category extension, and identifies general structuring principles that appear to operate across different semantic domains.

Abstract

The Max Planck Institute for Psycholinguistics (MPI) manages an archive of linguistic research data with a current size of almost 20 Terabytes. Apart from in-house researchers other projects also store their data in the archive, most notably the Documentation of Endangered Languages (DoBeS) projects. The archive is available online and can be accessed by anybody with Internet access. To be able to manage this large amount of data the MPI's technical group has developed a software suite called Language Archiving Technology (LAT) that on the one hand helps researchers and archive managers to manage the data and on the other hand helps users in enriching their primary data with additional layers. All the MPI software is Java-based and developed according to open source principles (GNU, 2007). All three major operating systems (Windows, Linux, MacOS) are supported and the software works similarly on all of them. As the archive is online, many of the tools, especially the ones for accessing the data, are browser based. Some of these browser-based tools make use of Adobe Flex to create nice-looking GUIs. The LAT suite is a complete set of management and enrichment tools, and given the interaction between the tools the result is a complete LAT software domain. Over the last 10 years, this domain has proven its functionality and use, and is being deployed to servers in other institutions. This deployment is an important step in getting the archived resources back to the members of the speech communities whose languages are documented. In the paper we give an overview of the tools of the LAT suite and we describe their functionality and role in the integrated process of archiving, management and enrichment of linguistic data.

Abstract

Mental imagery is a cognitive process crucial to human reasoning. Numerous studies have characterized specific
instances of this cognitive ability, as evoked by visual imagery (VI) or motor imagery (MI) tasks. However, it
remains unclear which neural resources are shared between VI and MI, and which are exclusively related to MI.
To address this issue, we have used fMRI to measure human brain activity during performance of VI and MI
tasks. Crucially, we have modulated the imagery process by manipulating the degree of mental rotation necessary
to solve the tasks. We focused our analysis on changes in neural signal as a function of the degree of mental
rotation in each task.

Abstract

Interpreting the meaning of co-speech gesture often involves
identifying a gesture’s ‘lexical affiliate’, the word or phrase to
which it most closely relates (Schegloff 1984). Though there is
work within gesture studies that resists this simplex mapping of
meaning from speech to gesture (e.g. de Ruiter 2000; Kendon
2014; Parrill 2008), including an evolving body of literature on
recurrent gesture and gesture families (e.g. Fricke et al. 2014; Müller 2017), it is still the lexical affiliate model that is most ap-
parent in formal linguistic models of multimodal meaning(e.g.
Alahverdzhieva et al. 2017; Lascarides and Stone 2009; Puste-
jovsky and Krishnaswamy 2021; Schlenker 2020). In this work,
I argue that the lexical affiliate should be carefully reconsidered
in the further development of such models.
In place of the lexical affiliate, I suggest a further shift
toward a frame-based, action schematic approach to gestural
meaning in line with that proposed in, for example, Parrill and
Sweetser (2004) and Müller (2017). To demonstrate the utility
of this approach I present three types of compositional gesture
sequences which I call spatial contrast, spatial embedding, and
cooperative abstract deixis. All three rely on gestural context,
rather than gesture-speech alignment, to convey interactive (i.e.
pragmatic) meaning. The centrality of gestural context to ges-
ture meaning in these examples demonstrates the necessity of
developing a model of gestural meaning independent of its in-
tegration with speech.

Abstract

Individual and interactive movement behaviour (non-verbal behaviour / communication) specifically reflects implicit processes in psychotherapy [1,4,11]. However, thus far, the registration of movement behaviour has been a methodological challenge. We will present a coding system combined with an annotation tool for the analysis of movement behaviour during psychotherapy interviews [9]. The NGCS coding system enables to classify body movements based on their kinetic features alone [5,7]. The theoretical assumption behind the NGCS is that its main kinetic and functional movement categories are differentially associated with specific psychological functions and thus, have different neurobiological correlates [5-8]. ELAN is a multimodal annotation tool for digital video media [2,3,12]. The NGCS / ELAN template enables to link any movie to the same coding system and to have different raters independently work on the same file. The potential of movement behaviour analysis as an objective tool for psychotherapy research and for supervision in the psychosomatic practice is discussed by giving examples of the NGCS/ELAN analyses of psychotherapy sessions. While the quality of kinetic turn-taking and the therapistrsquor;s (implicit) adoption of the patientrsquor;s movements may predict therapy outcome, changes in the patientrsquor;s movement behaviour pattern may indicate changes in cognitive concepts and emotional states and thus, may help to identify therapeutically relevant processes [10].

Abstract

This paper presents a novel model for 3D image segmentation and reconstruction. It has been designed with the aim to be implemented over a computer cluster or a multi-core platform. The required features include a nearly absolute independence between the processes participating in the segmentation task and providing amount of work as equal as possible for all the participants. As a result, it is avoid many drawbacks often encountered when performing a parallelization of an algorithm that was constructed to operate in a sequential manner. Furthermore, the proposed algorithm based on the new segmentation model is efficient and shows a very good, nearly linear performance growth along with the growing number of processing units.

Abstract

Deformable models are a popular family of image segmentation techniques, which has been gaining significant focus in the last two decades, serving both for real-world applications as well as the base for research work. One of the features that the deformable models offer and that is considered a much desired one, is the ability to change their topology during the segmentation process. Using this characteristic it is possible to perform segmentation of objects with discontinuities in their bodies or to detect an undefined number of objects in the scene. In this paper we present our model for handling the topology changes in image segmentation methods based on the Active Volumes solution. The said model is capable of performing the changes in the structure of objects while the segmentation progresses, what makes it efficient and suitable for implementations over powerful execution environment, like multi-core architectures or computer clusters.

Abstract

In this paper we present a novel approach for image segmentation using Active Nets and Active Volumes. Those solutions are based on the Deformable Models, with slight difference in the method for describing the shapes of interests - instead of using a contour or a surface they represented the segmented objects with a mesh structure, which allows to describe not only the surface of the objects but also to model their interiors. This is obtained by dividing the nodes of the mesh in two categories, namely internal and external ones, which will be responsible for two different tasks. In our new approach we propose to negate this separation and use only one type of nodes. Using that assumption we manage to significantly shorten the time of segmentation while maintaining its quality.

Abstract

In this paper we report on four experiments in which we attempted to prime the stress position of Dutch bisyllabic target nouns. These nouns, picture names, had stress on either the first or the second syllable. Auditory prime words had either the same stress as the target or a different stress (e.g., WORtel – MOtor vs. koSTUUM – MOtor; capital letters indicate stressed syllables in prime – target pairs). Furthermore, half of the prime words were semantically related, the other half were unrelated. In none of the experiments a stress priming effect was found. This could mean that stress is not stored in the lexicon. An additional finding was that targets with initial stress had a faster response than targets with a final stress. We hypothesize that bisyllabic words with final stress take longer to be encoded because this stress pattern is irregular with respect to the lexical distribution of bisyllabic stress patterns, even though it can be regular in terms of the metrical stress rules of Dutch.

Abstract

Large language models that exhibit instruction-following behaviour represent one of the biggest recent upheavals in conversational interfaces, a trend in large part fuelled by the release of OpenAI's ChatGPT, a proprietary large language model for text generation fine-tuned through reinforcement learning from human feedback (LLM+RLHF). We review the risks of relying on proprietary software and survey the first crop of open-source projects of comparable architecture and functionality. The main contribution of this paper is to show that openness is differentiated, and to offer scientific documentation of degrees of openness in this fast-moving field. We evaluate projects in terms of openness of code, training data, model weights, RLHF data, licensing, scientific documentation, and access methods. We find that while there is a fast-growing list of projects billing themselves as 'open source', many inherit undocumented data of dubious legality, few share the all-important instruction-tuning (a key site where human labour is involved), and careful scientific documentation is exceedingly rare. Degrees of openness are relevant to fairness and accountability at all points, from data collection and curation to model architecture, and from training and fine-tuning to release and deployment.

Abstract

Speech recognition systems are a key intermediary in voice-driven human-computer interaction. Although speech recognition works well for pristine monologic audio, real-life use cases in open-ended interactive settings still present many challenges. We argue that timing is mission-critical for dialogue systems, and evaluate 5 major commercial ASR systems for their conversational and multilingual support. We find that word error rates for natural conversational data in 6 languages remain abysmal, and that overlap remains a key challenge (study 1). This impacts especially the recognition of conversational words (study 2), and in turn has dire consequences for downstream intent recognition (study 3). Our findings help to evaluate the current state of conversational ASR, contribute towards multidimensional error analysis and evaluation, and identify phenomena that need most attention on the way to build robust interactive speech technologies.

Abstract

The human body is unique: it is both an object of perception and the source of human experience. Its universality makes it a perfect resource for asking questions about how cognition, brain and typology relate to one another. For example, we can ask how speakers of different languages segment and categorize the human body. A dominant view is that body parts are “given” by visual perceptual discontinuities, and that words are merely labels for these visually determined parts (e.g., Andersen, 1978; Brown, 1976; Lakoff, 1987). However, there are problems with this view. First it ignores other perceptual information, such as somatosensory and motoric representations. By looking at the neural representations of sesnsory representations, we can test how much of the categorization of the human body can be done through perception alone. Second, we can look at language typology to see how much universality and variation there is in body-part categories. A comparison of a range of typologically, genetically and areally diverse languages shows that the perceptual view has only limited applicability (Majid, Enfield & van Staden, in press). For example, using a “coloring-in” task, where speakers of seven different languages were given a line drawing of a human body and asked to color in various body parts, Majid & van Staden (in prep) show that languages vary substantially in body part segmentation. For example, Jahai (Mon-Khmer) makes a lexical distinction between upper arm, lower arm, and hand, but Lavukaleve (Papuan Isolate) has just one word to refer to arm, hand, and leg. This shows that body part categorization is not a straightforward mapping of words to visually determined perceptual parts.

Abstract

Many studies in cognitive science address how people categorize objects, but there has been comparatively little research on event categorization. This study investigated the categorization of events involving material destruction, such as “cutting” and “breaking”. Speakers of 28 typologically, genetically, and areally diverse languages described events shown in a set of video-clips. There was considerable cross-linguistic agreement in the dimensions along which the events were distinguished, but there was variation in the number of categories and the placement of their boundaries.

Abstract

Visual and auditory channels have different affordances and
this is mirrored in what information is available for linguistic
encoding. The visual channel has high spatial acuity, whereas
the auditory channel has better temporal acuity. These
differences may lead to different conceptualizations of events
and affect multimodal language production. Previous studies of
motion events typically present visual input to elicit speech and
gesture. The present study compared events presented as audio-
only, visual-only, or multimodal (visual+audio) input and
assessed speech and co-speech gesture for path and manner of
motion in Turkish. Speakers with audio-only input mentioned
path more and manner less in verbal descriptions, compared to
speakers who had visual input. There was no difference in the
type or frequency of gestures across conditions, and gestures
were dominated by path-only gestures. This suggests that input
modality influences speakers’ encoding of path and manner of
motion events in speech, but not in co-speech gestures.

Abstract

The goal of the presented study was to investigate the use of coarticulatory vowel nasalization in lexical access by native speakers of American English. In particular, we compare the use of coart culatory place of articulation cues to that of coarticulatory vowel nasalization. Previous research on lexical access has shown that listeners use cues to the place of articulation of a postvocalic stop in the preceding vowel. However, vowel nasalization as cue to an upcoming nasal consonant has been argued to be a more complex phenomenon. In order to establish whether coarticulatory vowel nasalization aides in the process of lexical access in the same way as place of articulation cues do, we conducted two perception experiments: an off-line 2AFC discrimination task and an on-line eyetracking study using the visual world paradigm. The results of our study suggest that listeners are indeed able to use vowel nasalization in similar ways to place of articulation information, and that both types of cues aide in lexical access.

Abstract

This thesis addressed the spoken production of complex numerals for time and space. The production of complex numerical expressions like those involved in telling time (e.g., 'quarter to four') or producing house numbers (e.g., 'two hundred forty-five') has been almost completely ignored. Yet, adult speakers produce such expressions on a regular basis in everyday communication. Thus, no theory on numerical cognition or speech production is complete without an account of the production of multi-morphemic utterances such as complex numeral expressions. The main question of this thesis is which particular speech planning levels are involved in the naming and reading of complex numerals for time and space. More specifically, this issue was investigated by examining different modes of response (clock times versus house numbers), alternative input formats (Arabic digit versus alphabetic format; analog versus digital clock displays), and different expression types (relative 'quarter to four' versus absolute 'three forty-five' time expressions).

Abstract

Recurrent neural networks (RNNs) have long been an architecture of interest for computational models of human sentence processing. The recently introduced Transformer architecture outperforms RNNs on many natural language processing tasks but little is known about its ability to model human language processing. We compare Transformer- and RNN-based language models’ ability to account for measures of human reading effort. Our analysis shows Transformers to outperform RNNs in explaining self-paced reading times and neural activity during reading English sentences, challenging the widely held idea that human sentence processing involves recurrent and immediate processing and provides evidence for cue-based retrieval.

Abstract

This study addresses the question whether visually grounded speech recognition (VGS) models learn to capture sentence semantics without access to any prior linguistic knowledge. We produce synthetic and natural spoken versions of a well known semantic textual similarity database and show that our VGS model produces embeddings that correlate well with human semantic similarity judgements. Our results show that a model trained on a small image-caption database outperforms two models trained on much larger databases, indicating that database size is not all that matters. We also investigate the importance of having multiple captions per image and find that this is indeed helpful even if the total number of images is lower, suggesting that paraphrasing is a valuable learning signal. While the general trend in the field is to create ever larger datasets to train models on, our findings indicate other characteristics of the database can just as important.

Abstract

Language emergence is characterized by a high degree of lex-
ical variation. It has been suggested that the speed at which
lexical conventionalization occurs depends partially on social
structure. In large communities, individuals receive input from
many sources, creating a pressure for lexical convergence.
In small, insular communities, individuals can remember id-
iolects and share common ground with interlocuters, allow-
ing these communities to retain a high degree of lexical vari-
ation. We look at lexical variation in Kata Kolok, a sign lan-
guage which emerged six generations ago in a Balinese vil-
lage, where women tend to have more tightly-knit social net-
works than men. We test if there are differing degrees of lexical
uniformity between women and men by reanalyzing a picture
description task in Kata Kolok. We find that women’s produc-
tions exhibit less lexical uniformity than men’s. One possible
explanation of this finding is that women’s more tightly-knit
social networks allow for remembering idiolects, alleviating
the pressure for lexical convergence, but social network data
from the Kata Kolok community is needed to support this ex-
planation.

Abstract

Corpus linguistics and language documentation are usually considered separate subdisciplines within linguistics, having developed from different traditions and often operating on different scales, but the authors will suggest that there are commonalities to the two: both aim to represent language use in a community, and both are concerned with managing digital data. The authors propose that the development of the Australian National Corpus (AusNC) be guided by the experience of language documentation in the management of multimodal digital data and its annotation, and in ethical issues pertaining to making the data accessible. This would allow an AusNC that is distributed, multimodal, and multilingual, with holdings of text, audio, and video data distributed across multiple institutions; and including Indigenous, sign, and migrant community languages. An audit of language material held by Australian institutions and individuals is necessary to gauge the diversity and volume of possible content, and to inform common technical standards.

Abstract

Many studies investigating the temporal alignment of co-speech
gestures to acoustic units in the speech signal find a close
coupling of the gestural landmarks and pitch accents or the
stressed syllable of pitch-accented words. In English, a pitch
accent is anchored in the lexically stressed syllable. Hence, it is
unclear whether it is the lexical phonological dimension of
stress, or the phrase-level prominence that determines the
details of speech-gesture synchronization. This paper explores
the relation between gestural phase transitions and accentual F0
peaks in Stockholm Swedish, which exhibits a lexical pitch
accent distinction. When produced with phrase-level
prominence, there are three different configurations of
lexicality of F0 peaks and the status of the syllable it is aligned
with. Through analyzing the alignment of the different F0 peaks
with gestural onsets in spontaneous dyadic conversations, we
aim to contribute to our understanding of the role of lexical
prosodic phonology in the co-production of speech and gesture.
The results, though limited by a small dataset, still suggest
differences between the three types of peaks concerning which
types of gesture phase onsets they tend to align with, and how
well these landmarks align with each other, although these
differences did not reach significance.

Abstract

Many Mayan languages make use of a special dependent verb form (the Agent Focus, or AF verb form), which alternates with the normal transitive verb form (the synthetic verb form) of main clauses when the subject of a transitive verb is focused, questioned or relativized. It has been a centerpiece of research in Mayan morphosyntax over the last forty years, due to its puzzling formal and distributional properties. In this dissertation I show how a usage-oriented approach to the phenomenon can provide important insights into this area of grammar which resists any categorical explanation. I propose that the historical origins of these special verb forms can be traced to the emergence of head marking. Drawing on cross-linguistic and historical data, I argue that the special verbs that occur in A-bar dependencies in Yucatec and a range of head-marking languages are byproducts of the frequency-sensitive gramaticalization process by which independent pronouns become pronominal inflection on verbs. I show that the relatively low frequency of adjacent pronoun-verb combinations in extraction contexts (where gaps are more frequent than resumptive pronouns) can give rise to asymmetric patterns of pronoun grammaticalization, and thus lead to the emergence of these morphological alternations. The asymmetric frequency distributions of gaps and RPs (within and across languages) in turn can be explained by processing preferences. I present three experiments which show that Yucatec speakers are more likely to use the resumptive verb form in embedded environments, and where the antecedent is indefinite. Specifically, these studies indicate the need to bring discourse-level processing principles into the account of what have often been taken to be autonomously sentence-internal phenomena: factors such as distance and the referential salience of the antecedent have been shown to influence referential form choice in discourse, suggesting that the same cognitive principles lie behind both types of variation. More generally, the Yucatec studies demonstrate that production preferences in Yucatec relative clauses reflect patterns of RP/gap distributions that have been attested across grammars. The Highest Subject Restriction (the ban on subject RPs in local dependencies), which is apparently a categorical constraint in many languages, is reflected probabilistically in Yucatec in terms of production preferences. The definiteness restriction (RPs are obligatory with indefinite antecedents), which has been reported categorically in other languages, is also visible probabilistically in Yucatec production. This lends some statistically robust support to the view that typological patterns can arise via the conventionalization of processing preferences.

Abstract

Typologies are critical tools for linguists, but typologies, like grammars, are known to leak. This book addresses the question of typological overlap from the perspective of a single language. In Lowland Chontal of Oaxaca, a language of southern Mexico, change events are expressed with three types of predicates, and each predicate type corresponds to a different language type in the well-known typology of lexicalization patterns established by Talmy and elaborated by others. O’Connor evaluates the predictive powers of the typology by examining the consequences of each predicate type in a variety of contexts, using data from narrative discourse, stimulus response, and elicitation. This is the first de­tailed look at the lexical and grammatical resources of the verbal system in Chontal and their relation to semantics of change. The analysis of how and why Chontal speakers choose among these verbal resources to achieve particular communicative and social goals serves both as a documentation of an endangered language and a theoretical contribution towards a typology of language use.

Abstract

Phonetic convergence—i.e., adapting one’s speech
towards that of an interlocutor—has been shown
to occur in human-human conversations as well as
human-machine interactions. Here, we investigate
the hypothesis that human-to-robot convergence is
influenced by the human’s perception of the robot
and by the conversation’s topic. We conducted a
within-subjects experiment in which 33 participants
interacted with two robots differing in their eye gaze
behavior—one looked constantly at the participant;
the other produced gaze aversions, similarly to a
human’s behavior. Additionally, the robot asked
questions with increasing intimacy levels.
We observed that the speakers tended to converge
on F0 to the robots. However, this convergence
to the robots was not modulated by how the
speakers perceived them or by the topic’s intimacy.
Interestingly, speakers produced lower F0 means
when talking about more intimate topics. We
discuss these findings in terms of current theories of
conversational convergence.

Abstract

Japanese, British English and American English listeners were presented with spoken words in their native language, and asked to mark on a written transcript of each word the first natural division point in the word. The results showed clear and strong patterns of consensus, indicating that listeners have available to them conscious representations of within-word structure. Orthography did not play a strongly deciding role in the results. The patterns of response were at variance with results from on-line studies of speech segmentation, suggesting that the present task taps not those representations used in on-line listening, but levels of representation which may involve much richer knowledge of word-internal structure.

Abstract

In this paper we describe a general scheme for internationalization of our talking head, Baldi, to speak other languages. We describe the modular structure of the auditory/visual synthesis software. As an example, we have created a synthetic Arabic talker, which is evaluated using a noisy word recognition task comparing this talker with a natural one.

Abstract

LEXUS and ViCoS: from lexicon to conceptual spaces LEXUS is a web-based lexicon tool and the knowledge space software ViCoS is an extension of LEXUS, allowing users to create relations between objects in and across lexica. LEXUS and ViCoS are part of the Language Archiving Technology software, developed at the MPI for Psycholinguistics to archive and enrich linguistic resources collected in the framework of language documentation projects. LEXUS is of primary interest for language documentation, offering the possibility to not just create a digital dictionary, but additionally it allows the creation of multi-media encyclopedic lexica. ViCoS provides an interface between the lexical space and the ontological space. Its approach permits users to model a world of concepts and their interrelations based on categorization patterns made by the speech community. We describe the LEXUS and ViCoS functionalities using three cases from DoBeS language documentation projects: (1) Marquesan The Marquesan lexicon was initially created in Toolbox and imported into LEXUS using the Toolbox import functionality. The lexicon is enriched with multi-media to illustrate the meaning of the words in its cultural environment. Members of the speech community consider words as keys to access and describe relevant parts of their life and traditions. Their understanding of words is best described by the various associations they evoke rather than in terms of any formal theory of meaning. Using ViCoS a knowledge space of related concepts is being created. (2) Kola-Sámi Two lexica are being created in LEXUS: RuSaDic lexicon is a Russian-Kildin wordlist in which the entries are of relative limited structure and content. SaRuDiC is a more complex structured lexicon with much richer content, including multi-media fragments and derivations. Using ViCoS we have created a connection between the two lexica, so that speakers who are familiair with Russian and wish to revitalize their Kildin can enter the lexicon through the RuSaDic and from there approach the informative SaRuDic. Similary we will create relations from the two lexica to external open databases, like e.g. Álgu. (3) Beaver A speaker database including kinship relations has been created and the database has been imported into LEXUS. In the LEXUS views the relations for individual speakers are being displayed. Using ViCoS the relational information from the database will be extracted to form a kisnhip relation space with specific relation types, like e.g 'mother-of'. The whole set of relations from the database can be displayed in one ViCoS relation window, and zoom functionality is available.

Abstract

In this paper, I will argue that although the study of multimodal interaction offers exciting new prospects for Human Computer Interaction and human-human communication research, language is the primary form of communication, even in multimodal systems. I will support this claim with theoretical and empirical arguments, mainly drawn from human-human communication research, and will discuss the implications for multimodal communication research and Human-Computer Interaction.

Abstract

Children’s ability to share attention with another social partner (i.e., joint attention) has been found to support language development. Despite the large amount of research examining the effects of joint attention on language in hearing population, little is known about how deaf children learning sign languages achieve joint attention with their caregivers during natural social interaction and how caregivers provide and scaffold learning opportunities for their children. The present study investigates the properties and timing of joint attention surrounding familiar and novel naming events and their relationship to children’s vocabulary. Naturalistic play sessions of caretaker-child-dyads using American Sign Language were analyzed in regards to naming events of either familiar or novel object labeling events and the surrounding joint attention events. We observed that most naming events took place in the context of a successful joint attention event and that sign familiarity was related to the timing of naming events within the joint attention events. Our results suggest that caregivers are highly sensitive to their child’s visual attention in interactions and modulate joint attention differently in the context of naming events of familiar vs. novel object labels.

Abstract

Most of the study of basic emotion expressions has focused on facial expressions and little work has been done to specifically investigate happiness, the only positive of the basic emotions (Ekman & Friesen, 1971). However, a theoretical suggestion has been made that happiness could be broken down into discrete positive emotions, which each fulfil the criteria of basic emotions, and that these would be expressed vocally (Ekman, 1992). To empirically test this hypothesis, 20 participants categorised 80 paralinguistic sounds using the labels achievement, amusement, contentment, pleasure and relief. The results suggest that achievement, amusement and relief are perceived as distinct categories, which subjects accurately identify. In contrast, the categories of contentment and pleasure were systematically confused with other responses, although performance was still well above chance levels. These findings are initial evidence that the positive emotions engage distinct vocal expressions and may be considered to be distinct emotion categories.

Abstract

Emotional signals allow for the sharing of important information with conspecifics, for example to warn them of danger. Humans use a range of different cues to communicate to others how they feel, including facial, vocal, and gestural signals. Although much is known about facial expressions of emotion, less research has focused on affect in the voice. We compare British listeners to individuals from remote Namibian villages who have had no exposure to Western culture, and examine recognition of non-verbal emotional vocalizations, such as screams and laughs. We show that a number of emotions can be universally recognized from non-verbal vocal signals. In addition we demonstrate the specificity of this pattern, with a set of additional emotions only recognized within, but not across these cultural groups. Our findings indicate that a small set of primarily negative emotions have evolved signals across several modalities, while most positive emotions are communicated with culture-specific signals.

Abstract

In this paper, we investigate the ability of SpeM, our recognition system based on the combination of an automatic phone recogniser and a wordsearch module, to determine as early as possible during the word recognition process whether a word is likely to be recognised correctly (this we refer to as ‘on-line’ early word recognition). We present two measures that can be used to predict whether a word is correctly recognised: the Bayesian word activation and the amount of available (acoustic) information for a word. SpeM was tested on 1,463 polysyllabic words in 885 continuous speech utterances. The investigated predictors indicated that a word activation that is 1) high (but not too high) and 2) based on more phones is more reliable to predict the correctness of a word than a similarly high value based on a small number of phones or a lower value of the word activation.

Abstract

We have recently developed a new model of human speech recognition, based on automatic speech recognition techniques [1]. The present paper has two goals. First, we show that the new model performs well in the recognition of lexically ambiguous input. These demonstrations suggest that the model is able to operate in the same optimal way as human listeners. Second, we discuss how to relate the behaviour of a recogniser, designed to discover the optimum path through a word lattice, to data from human listening experiments. We argue that this requires a metric that combines both path-based and word-based measures of recognition performance. The combined metric varies continuously as the input speech signal unfolds over time.

Abstract

A stretch of speech is often consistent with multiple words, e.g., the sequence /hæm/ is consistent with ‘ham’ but also with the first syllable of ‘hamster’, resulting in temporary ambiguity. However, to what degree does this lexical embedding occur? Analyses on two corpora of spoken Dutch showed that 11.9%-19.5% of polysyllabic word tokens have word-initial embedding, while 4.1%-7.5% of monosyllabic word tokens can appear word-initially embedded. This is much lower than suggested by an analysis of a large dictionary of Dutch. Speech processing thus appears to be simpler than one might expect on the basis of statistics on a dictionary.

Abstract

In this paper, we present a novel computational model of human speech recognition – called SpeM – based on the theory underlying Shortlist. We will show that SpeM, in combination with an automatic phone recogniser (APR), is able to simulate the human speech recognition process from the acoustic signal to the ultimate recognition of words. This joint model takes an acoustic speech file as input and calculates the activation flows of candidate words on the basis of the degree of fit of the candidate words with the input. Experiments showed that SpeM outperforms Shortlist on the recognition of ‘real-life’ input. Furthermore, SpeM performs only slightly worse than an off-the-shelf full-blown automatic speech recogniser in which all words are equally probable, while it provides a transparent computationally elegant paradigm for modelling word activations in human word recognition.

Abstract

Evidence that listeners use durational cues to help resolve temporarily ambiguous speech input has accumulated over the past few years. In this paper, we investigate whether durational cues are also beneficial for word recognition in a computational model of spoken-word recognition. Two sets of simulations were carried out using the acoustic signal as input. The simulations showed that the computational model, like humans, takes benefit from durational cues during word recognition, and uses these to disambiguate the speech signal. These results thus provide support for the theory that durational cues play a role in spoken-word recognition.

Abstract

This study investigated the encoding of metrical information during speech production in Dutch. In Experiment 1, participants were asked to judge whether bisyllabic picture names had initial or final stress. Results showed significantly faster decision times for initially stressed targets (e.g., LEpel 'spoon') than for targets with final stress (e.g., liBEL 'dragon fly'; capital letters indicate stressed syllables) and revealed that the monitoring latencies are not a function of the picture naming or object recognition latencies to the same pictures. Experiments 2 and 3 replicated the outcome of the first experiment with bi- and trisyllabic picture names. These results demonstrate that metrical information of words is encoded rightward incrementally during phonological encoding in speech production. The results of these experiments are in line with Levelt's model of phonological encoding.

Abstract

Sarah Schimke onderzocht hoe mensen die op volwassen leeftijd naar een ander land verhuizen de taal van dit land leren, ook zonder veel taalinstructie te krijgen. Twee groepen werden onderzocht: Turkse immigranten in Frankrijk en Turkse immigranten in Duitsland. De resultaten laten zien dat volwassen leerlingen in het begin van het verwervingsproces een gemakkelijkere variatie van de doeltaal creëren. Er worden wel woorden van de doeltaal verworven en gebruikt, maar er wordt een gesimplificeerde grammatica toegepast. In het bijzonder gebruiken leerlingen in deze fase geen finietheid, dus geen morfologische variaties van werkwoorden. Schimke toont aan dat als finietheid wordt verworven, dit de grammatica van de leerlingen sterk verandert en dat deze veel sterker op de doeltaalgrammatica begint te lijken. Ook toont ze aan dat dit proces door karakteristieken van de doeltaal, zoals de woordvolgorde en de complexiteit van de morfologie, wordt beïnvloed

Abstract

This paper presents a study on the reduction of word-final [t]s in conversational standard Dutch. Based on a large amount of tokens annotated on the segmental level, we show that the bigram frequency and the segmental context are the main predictors for the absence of [t]s. In a second study, we present an analysis of the detailed acoustic properties of word-final [t]s and we show that bigram frequency and context also play a role on the subsegmental level. This paper extends research on the realization of /t/ in spontaneous speech and shows the importance of incorporating sub-segmental properties in models of speech.

Abstract

Previous studies have indicated that vocal and facial expressions of the ‘basic’ emotions share aspects of processing. Thus amygdala damage compromises the perception of fear and anger from the face and from the voice. In the current study we tested the hypothesis that there exist positive basic emotions, expressed mainly in the voice (Ekman, 1992). Vocal stimuli were produced to express the specific positive emotions of amusement, achievement, pleasure, contentment and relief.

Abstract

This study investigated the impact of a speaker's gaze direction
on a listener's comprehension of discourse. Previous research
suggests that hand gestures play a role in referent allocation,
enabling listeners to better understand the discourse. The
current study aims to determine whether the speaker's gaze
direction has a similar effect on reference resolution as co-
speech gestures. Thirty native Japanese speakers participated in
the study and were assigned to one of three conditions:
congruent, incongruent, or speech-only. Participants watched
36 videos of an actor narrating a story consisting of three
sentences with two protagonists. The speaker consistently
used hand gestures to allocate one protagonist to the lower right
and the other to the lower left space, while directing her gaze to
either space of the target person (congruent), the other person
(incongruent), or no particular space (speech-only). Participants
were required to verbally answer a question about the target
protagonist involved in an accidental event as quickly as
possible. Results indicate that participants in the congruent
condition exhibited faster reaction times than those in the
incongruent condition, although the difference was not
significant. These findings suggest that the speaker's gaze
direction is not enough to facilitate a listener's comprehension
of discourse.

Abstract

The present paper is part of a large research programme investigating the nature and properties of the predicate logic inherent in natural language. The general hypothesis is that natural speakers start off with a basic-natural logic, based on natural cognitive functions, including the basic-natural way of dealing with plural objects. As culture spreads, functional pressure leads to greater generalization and mathematical correctness, yielding ever more refined systems until the apogee of standard modern predicate logic. Four systems of predicate calculus are considered: Basic-Natural Predicate Calculus (BNPC), Aritsotelian-Abelardian Predicate Calculus (AAPC), Aritsotelian-Boethian Predicate Calculus (ABPC), also known as the classic Square of Opposition, and Standard Modern Predicate Calculus (SMPC). (ABPC is logically faulty owing to its Undue Existential Import (UEI), but that fault is repaired by the addition of a presuppositional component to the logic.) All four systems are checked against seven natural logical intuitions. It appears that BNPC scores best (five out of seven), followed by ABPC (three out of seven). AAPC and SMPC finish ex aequo with two out of seven.

Abstract

Speech is perceived relative to the speech rate in the context. It is unclear, however, what information listeners use to compute speech rate. The present study examines whether listeners use the number of
syllables per unit time (i.e., syllable rate) as a measure of speech rate, as indexed by subsequent vowel perception. We ran two rate-normalization experiments in which participants heard duration-matched word lists that contained either monosyllabic
vs. bisyllabic words (Experiment 1), or monosyllabic vs. trisyllabic pseudowords (Experiment 2). The participants’ task was to categorize an /ɑ-aː/ continuum that followed the word lists. The monosyllabic condition was perceived as slower (i.e., fewer /aː/ responses) than the bisyllabic and
trisyllabic condition. However, no difference was observed between bisyllabic and trisyllabic contexts. Therefore, while syllable rate is used in perceiving speech rate, other factors, such as fast speech processes, mean F0, and intensity, must also influence rate normalization.