Domain Of Speech Research Articles

Collaboration and networking in research grant projects in speech, language and hearing sciences

This study analysed collaboration and networking in research grant projects of the research fund of the All India Institute of Speech and Hearing (ARF) during 2001–02 to 2015–16 in 2 spans of seven years. The recent span of seven years witnessed higher collaboration with higher productivity. While the number of projects in the domain of speech was the highest, followed by multidisciplinary in span I, the projects in the domain of language was the highest, followed by hearing in span II. The number of local collaborations were the highest. The domestic and international collaborations witnessed an increase in the recent years. Greater networking was seen between junior and senior faculty. This study adds on to the existing literature on social system perspective of science to understand the dynamics of networking pattern which can provide meaningful insights to formulate effective strategies to enhance research productivity in this discipline.

When does speech sound disorder matter for literacy? The role of disordered speech errors, co-occurring language impairment and family risk of dyslexia.

BackgroundThis study considers the role of early speech difficulties in literacy development, in the context of additional risk factors.MethodChildren were identified with speech sound disorder (SSD) at the age of 3½ years, on the basis of performance on the Diagnostic Evaluation of Articulation and Phonology. Their literacy skills were assessed at the start of formal reading instruction (age 5½), using measures of phoneme awareness, word‐level reading and spelling; and 3 years later (age 8), using measures of word‐level reading, spelling and reading comprehension.ResultsThe presence of early SSD conferred a small but significant risk of poor phonemic skills and spelling at the age of 5½ and of poor word reading at the age of 8. Furthermore, within the group with SSD, the persistence of speech difficulties to the point of school entry was associated with poorer emergent literacy skills, and children with ‘disordered’ speech errors had poorer word reading skills than children whose speech errors indicated ‘delay’. In contrast, the initial severity of SSD was not a significant predictor of reading development. Beyond the domain of speech, the presence of a co‐occurring language impairment was strongly predictive of literacy skills and having a family risk of dyslexia predicted additional variance in literacy at both time‐points.ConclusionsEarly SSD alone has only modest effects on literacy development but when additional risk factors are present, these can have serious negative consequences, consistent with the view that multiple risks accumulate to predict reading disorders.

Collaboration trend on Speech, Language and Hearing Sciences: A Scientometric Study Based on Select Journals

<div><p><em>Collaboration trend among researchers can take many forms. One of the most tangible forms of this collaboration is co-authorship. Over the</em><em> decades the multiple-author publication, frequently referred to as a co-authored publication, has been used as a basic counting unit to measure collaborative activity. In order to identify the collaborative trend in speech, language and hearing science three leading journals such as Journal of Speech, Language and Hearing Research, published by American Speech and Hearing Association (ASHA), Asia Pacific Journal of Speech, Language and Hearing and Journal of the All India Institute of Speech and Hearing (JAIISH). A total of 905 articles were published in these three journals during the period of 2009-2013 has been considered for the study. Scienctometric tools such as Authorship pattern Collaborative Author Index, Degree of Collaboration. Collaboration Index, Collaborative Coefficient and Modified Collaborative Coefficient, Local collaborative index, Domestic collaborative index, International collaborative Index were used. The study shows that there exists collaborative research in speech, language and hearing sciences journals irrespective of the origin of the journal. The local collaboration of authors was more rather than domestic and international collaboration among the authors. The study also shows that there is no significant difference in collaboration in the domain of speech or language or hearing and Local collaboration persist in the domain of speech or language or hearing.</em></p></div>

Decoding social media speak: developing a speech act theory research agenda

Purpose– Drawing on the theoretical domain of speech act theory (SAT) and a discussion of its suitability for setting the agenda for social media research, this study aims to explore a range of research directions that are both relevant and conceptually robust, to stimulate the advancement of knowledge and understanding of online verbatim data.Design/methodology/approach– Examining previously published cross-disciplinary research, the study identifies how recent conceptual and empirical advances in SAT may further guide the development of text analytics in a social media context.Findings– Decoding content and function word use in customers’ social media communication can enhance the efficiency of determining potential impacts of customer reviews, sentiment strength, the quality of contributions in social media, customers’ socialization perceptions in online communities and deceptive messages.Originality/value– Considering the variety of managerial demand, increasing and diverging social media formats, expanding archives, rapid development of software tools and fast-paced market changes, this study provides an urgently needed, theory-driven, coherent research agenda to guide the conceptual development of text analytics in a social media context.

Audio-visual emotion recognition using multi-directional regression and Ridgelet transform

In this paper, we propose an audio-visual emotion recognition system using multi-directional regression (MDR) audio features and ridgelet transform based face image features. MDR features capture directional derivative information in a spectro-temporal domain of speech, and, thereby, suitable to encode different levels of increasing or decreasing pitch and formant frequencies. For video inputs, interest points in a time frame are detected using spectro-temporal filters, and ridgelet transform is applied to cuboids around the interest points. Two separate extreme learning machine classifiers, one for speech modality and the other for face modality, are used. The scores of these two classifiers are fused using a Bayesian sum rule to make the final decision. Experimental results on eNTERFACE database show that the proposed method achieves accuracy of 85.06 % using bimodal inputs, 64.04 % using speech only, and 58.38 % using face only; these accuracies outnumber the accuracies obtained by some other state-of-the-art systems using the same database.

Single-channel Dereverberation for Distant-Talking Speech Recognition by Combining Denoising Autoencoder and Temporal Structure Normalization

In this paper, we propose a robust distant-talking speech recognition by combining cepstral domain denoising autoencoder (DAE) and temporal structure normalization (TSN) filter. For the proposed method, after applying a DAE in the cepstral domain of speech to suppress reverberation, we apply a post-processing technology based on temporal structure normalization (TSN) filter to reduce the noise and reverberation effects by normalizing the modulation spectra to reference spectra of clean speech. The proposed method was evaluated using speech in simulated and real reverberant environments. By combining a cepstral-domain DAE and TSN, the average Word Error Rate (WER) was reduced from 25.2% of the baseline system to 21.2% in simulated environments and from 47.5% to 41.3% in real environments, respectively.

Politeness is Distance — Serbian Requests Through the Mirror of the Cognitive Speech Act Scenario Model

Our analysis was based on the theoretical model of speech act scenarios proposed by Panther & Thornburg (1998, 2003, 2004). We identified the components of the scenario in Serbian requests, showing that speakers most frequently resort to the BEFORE component in formulations of indirect requests implying the ability/capacity and willingness of the hearer to perform the act, with only sporadic instances of the RESULT and AFTER components. When it comes to the CORE component, our results show that Serbian possesses various linguistic means to express direct requests, in the first place by performative formulas, apart from the imperative proper. Further on, we make a synthesis of the two principal approaches within cognitive linguistics, the scenario approach and the propositional ICM approach, in order to account for the politeness effects produced by the mutual influence of sociocultural variables and the linguistic content of the utterance. Also, we observed certain inconsistencies in the application of the cognitive models, particularly in the domain of non-conventionalised indirect speech acts.

Total variability modelling for face verification

This study presents the first detailed study of total variability modelling (TVM) for face verification. TVM was originally proposed for speaker verification, where it has been accepted as state-of-the-art technology. Also referred to as front-end factor analysis, TVM uses a probabilistic model to represent a speech recording as a low-dimensional vector known as an ` i -vector'. This representation has been successfully applied to a wide variety of speech-related pattern recognition applications, and remains a hot topic in biometrics. In this work, the authors extend the application of i -vectors beyond the domain of speech to a novel representation of facial images for the purpose of face verification. Extensive experimentation on several challenging and publicly available face recognition databases demonstrates that TVM generalises well to this modality, providing between 17 and 39% relative reduction in verification error rate compared to a baseline Gaussian mixture model system. Several i -vector session compensation and scoring techniques were evaluated including source-normalised linear discriminant analysis (SN-LDA), probabilistic LDA and within-class covariance normalisation. Finally, this study provides a detailed comparison of the complexity of TVM, highlighting some important computational advantages with respect to related state-of-the-art techniques.

Attention modulates ‘speech-tracking’ at a cocktail party

Recent findings by Mesgarani and Chang demonstrate that signals in auditory cortex can reconstruct the spectrotemporal patterns of attended speech tokens better than those of ignored ones. These results help extend the study of attention into the domain of natural speech, posing numerous questions and challenges for future research.

Hebrew and Arabic in Asymmetric Contact in Israel

Hebrew and Arabic in Asymmetric Contact in Israel Israeli Hebrew (IH) and Palestinian Arabic (PA) 1 have existed side by side for well over a century in extremely close contact, accompanied by social and ideological tension, often conflict, between two communities: PA speakers, who turned from a majority to a minority following the establishment of the State of Israel in 1948, and IH speakers, the contemporary majority, representing the dominant culture. The Hebrew-speaking Jewish group is heterogeneous in terms of lands of origin and includes a large community of Judeo-Arabic speakers. In this paper I discuss the ‘when,’ ‘where,’ how,' and especially ‘why’ of mutual borrowing 2 between Hebrew and Arabic in Israel since the 1880s. I stress that the contact is asymmetric. The majority of Hebrew speakers today do not speak Arabic. They do, however, have access to borrowed lexicon in the domain of slang and vernacular speech that both enriches these registers and provides the speakers with a feeling of being ‘indigenous’ and ‘belonging’ to the social and cultural milieu of the larger environment. In contrast, most young adult PA-speakers are bilingual. For this minority, borrowing fulfills fundamental sociolinguistic and pragmatic needs. IH lexical items have permeated all areas of professional, educational, administrative, and social activities in everyday life discourse in all styles and registers of the spoken language and, to a lesser extent, written language. They fill lexical gaps and lacunae representing modern and western concepts that did not exist in PA. They also symbolize modernism, education, and social mobility. In other words, borrowing serves as a means of ‘ stepping in ’ for the Jewish majority and conversely as a means for ‘ stepping out and up ’ for the Arab minority.

Area Spt in the Human Planum Temporale Supports Sensory-Motor Integration for Speech Processing

Processing incoming sensory information and transforming this input into appropriate motor responses is a critical and ongoing aspect of our moment-to-moment interaction with the environment. While the neural mechanisms in the posterior parietal cortex (PPC) that support the transformation of sensory inputs into simple eye or limb movements has received a great deal of empirical attention-in part because these processes are easy to study in nonhuman primates-little work has been done on sensory-motor transformations in the domain of speech. Here we used functional magnetic resonance imaging and multivariate analysis techniques to demonstrate that a region of the planum temporale (Spt) shows distinct spatial activation patterns during sensory and motor aspects of a speech task. This result suggests that just as the PPC supports sensorimotor integration for eye and limb movements, area Spt forms part of a sensory-motor integration circuit for the vocal tract.

Race, Violence, and Terror: The Cultural Defensibility of Heteromasculine Citizenship in the Virginia Tech Massacre and the Don Imus Affair

Race, Violence, and TerrorThe Cultural Defensibility of Heteromasculine Citizenship in the Virginia Tech Massacre and the Don Imus Affair amy l. brandzel (bio) and jigna desai (bio) You had everything you wanted. Your Mercedes wasn't enough, you brats. Your golden necklaces weren't enough, you snobs. Your trust fund wasn't enough. Your vodka and cognac weren't enough. All your debaucheries weren't enough. Those weren't enough to fulfill your hedonistic needs. You had everything. Seung-hui Cho "Today, We Are All Hokies"1 perhaps the shock jock, Don Imus, did not shock many when his white male appeal seemed to digress too neatly into the domain of blatant racist speech. The sexual/racial/gender-loaded phrases by which he characterized the Rutgers University women's basketball team saturated the media, as commentators repeatedly asked along the way whether it was or was not acceptable for him to say what they said over and over. The phrase, not needed here, was hotly defended by numerous Imus supporters, especially through the deployment of the color-blindness rhetorical maneuver of equivalency. In the process, the violence of Imus's speech was removed from the field of white masculinity and quickly displaced onto African American communities and rap music: if African American rap artists can say it, why can't Don Imus?2 But, the Don Imus affair quickly receded into the back pages as another event unfolded: the "Virginia Tech massacre," as [End Page 61] it soon became called.3 The defense here was more subtle as Asian/American groups, and Korean/Americans in particular, were held (in/directly) accountable for the redeployment of another age, old dichotomy: the unassimilated Asian and the American citizen. While it might appear to be an odd pairing, we want to use the Don Imus affair as a lens through which to examine the privileges of normative citizenship and the "cultural defenses" that circulate in regards to violence in the United States. Put differently, we want to ask, how does the juxtaposition of Don Imus and Seung-hui Cho4 allow us to see the ways in which certain citizen-subjects are afforded and sanctioned the "right" to violence?5 How might we use this opportunity to interrogate the reoccurring appeal of white wounded masculinity as a "cultural defense" for violence, a violence borne from unobtainable heteronormative ideals? While the connection was rarely made obvious, what made Seung-hui Cho's actions palatable were the ways in which the media worked to squeeze him into the wounded-masculinity narrative: he fit the "type" for young male school shooters. He was a loner, a nerd, and a young male ostracized from the community due to his inability to access male privilege, social capital, confidence, and, most importantly, women. In many ways, this might merely be another way to describe the historically produced stereotype of Asian American masculinity, yet increasingly this "type" is also becoming the description of white middle-class suburban boys who are unable to utilize the properties of white heteronormative masculinity and who are increasingly becoming framed as the most wounded victims within and of the nation. But as much as Cho could be made to fit within this typecasting, he refused such analysis. Through his series of videos, polemics, and photos, Cho highlighted the ways in which his isolation was directly related to normative white citizenship, the alienation of Asian Americans, and disenfranchised racialized "queer" masculinities. Therefore, we suggest that perhaps the media, and white America in general, worked so hard to fit Cho within the wounded-masculinity type in order to avoid the other hermeneutical option: the racially oppressed retaliating for their isolation from the privileges of normative citizenship. [End Page 62] Don Imus: In Defense of the Good Ole Boys White heteromasculinity is a pervasive discourse that depends upon the presumption that it is and always will be "on the defense." Moreover, as a discourse, it is not directly tied to particular types of bodies—rather, anyone can participate and further white heteromasculinity by ventriloquising the narrative of the loss of privilege of masculinity and/or whiteness at the hands of feminists, progressives, academics, and racially aware others...

The status of purpose, reason, and intended endpoint in the typology of complex sentences: implications for layered models of clause structure

This study argues that constructions of purpose, reason and intended endpoint have a special status in typologies of complex sentences, because they combine features of the adverbial domain and the complement domain. The feature they share with adverbial clauses is that they do not function as the object of the predicate of the main clause, but instead elaborate on its circumstances. The feature they share with complement clauses is that unlike other adverbial clauses they specify the mental state of some participant in the main clause, typically the agent. Evidence for this special status comes from two sources. On the one hand, it is shown that the recognition of a mental state relation is necessary in semantic terms, to distinguish the three constructions from semantically similar types in the adverbial domain, and to explain their exceptional behavior in terms of presupposition. On the other hand, it is also shown that the presence of such a mental state relation is crosslinguistically reflected in the formal make-up of the three construction types, specifically in the use of mood markers in the dependent clause, and in the use of conjunctions that are shared with the domain of represented speech. From a theoretical perspective, the intermediate status of constructions of purpose, reason and intended endpoint is important because it provides structural evidence for layered models of clause structure, specifically for the distinction between a description-related and an interaction-related layer of organization, and for a further subdivision of the interaction-related layer in terms of modal and illocutionary features.

Holding Environment as Home

Holding Environment as Home

Karagoz Co-Opted: Turkish Shadow Theatre of the Early Republic (1923-1945)

During the early Turkish republic karagöz shadow theatre's Ottoman patterns were transformed. Reformers attempted to restrict coffeehouses where the art had flourished and developed written texts to replace the improvised practice of the past. They sponsored performances in government-supported community centers and created shows to promote government policies. The efforts meant that an art, which had grown from lower-class satire of the elite was purged of obscene elements, characters were changed to conform to modern ideology, and government control was asserted on what had been a domain of free speech. It is possible that the efforts to restructure the once vibrant art helped hasten its decline.

On-line Gaussian mixture modeling in the log-power domain for signal-to-noise ratio estimation and speech enhancement

We present on-line Gaussian mixture modeling (GMM) in the log-power domain of actual noisy speech and its applications to segmental signal-to-noise ratio (SNR) estimation and speech enhancement. The basic idea in this method is the use of conventional two-component GMM modeling in the log-power domain to estimate the distributions of noise and noisy speech subspaces in each speech segment of a length of 0.5–2 s. Given the subspace distributions, the statistical estimation method is adopted in the applications. For the segmental SNR estimation, the average speech level is estimated from noisy speech using a nonlinear moment of modeled distributions. This method is suitable under real conditions, when neither reference signals nor speech activity is available, and is shown to be more robust and accurate than conventional methods, particularly under low-SNR conditions. The proposed GMM model is extended to the multiband log-power domains for noise estimation. We use long-term information, which is obtained by GMM modeling in each segment of 0.5 s, to update the local distributions of noise and noisy speech power at each actual time–frequency index. The cumulative distribution function equalization (CDFE) is then used to estimate the noise and subtract it from the noisy speech power. The advantage of the CDFE method for noise estimation is that the estimation is given in the logarithmic domain without any approximation. The proposed speech enhancement is tested using the AURORA-2J database. We also compare the proposed method to the conventional minimum statistic and quantile-based noise estimation. The proposed method is found to be superior to the conventional in the speech recognition rate over most noise environments and shown to provide very good compromise between speech enhancement and speech recognition performance.

Sliding Doors—Opening Our World

In this article I share stories I have lived alongside my son, Matthew, who is challenged by multiple disabilities, particularly in the domain of expressive speech. Narrative inquiry shaped a space to attend and inquire into stories. The stories reveal the tension between an identity given and an identity continually created in relationship. Specifically the stories show that the “fix and serve” tradition in the education of children with disabilities can blank out attention to identity formation and agency of persons with disabilities who are making their way in the world as we all do. The stories offer consideration of the significance of imaginative relational play and improvisation for shaping participatory spaces from which evolving identities are continually created. I narrate stories, configured from my perspective as Matthew's mother, and I imagine ways to reshape ever more responsive educative landscapes for people with disabilities and for those in relation to them.

Superaddressee or Who Will Succeed a Mentor?

This philosophical essay is inspired by a four-year pedagogical relationship that continues in its altered form today. The main focus of this piece is the transformation of a mentor as an immediate addressee into mentor as a superaddressee, an influential third listener who oversees observable dialogues. I explore the mutual responsibilities of a student and a mentor in order to uncover the elements in the pedagogical chemistry responsible for the transformation of an addressee into a superaddressee. Confirmation (a perfect form of understanding) of a student’s intellectual and moral uniqueness, incarnation of a particular value deemed desirable by a student, and education of a student into the dialogic ways of being on his or her own are the necessary ingredients of the process of becoming a superaddressee. Initially a mentor engages in the pedagogy of understanding whose ideal outcome is confirmation of a student’s intellectual and moral makeup. After the dialogue is over, a mentor often moves into the domain of inner speech from where he or she continues to offer perfect understanding, especially in the absence of such understanding from an immediate addressee. Two types of superaddressees are identified and their relationship with the invoking consciousness is explored. I conclude that becoming a superaddressee is the most generous pedagogical contribution to a student’s future: a mentor thus makes his or her voice available to a student’s inner dialogue, often without receiving anything in return.

Method of noise reduction using correction and scaling vectors with partitioning of the acoustic space in the domain of noisy speech

A method and apparatus are provided for reducing noise in a training signal and/or test signal. The noise reduction technique uses a stereo signal formed of two channel signals, each channel containing the same pattern signal. One of the channel signals is “clean” and the other includes additive noise. Using feature vectors from these channel signals, a collection of noise correction and scaling vectors is determined. When a feature vector of a noisy pattern signal is later received, it is multiplied by the best scaling vector for that feature vector and the best correction vector is added to the product to produce a noise reduced feature vector. Under one embodiment, the best scaling and correction vectors are identified by choosing an optimal mixture component for the noisy feature vector. The optimal mixture component being selected based on a distribution of noisy channel feature vectors associated with each mixture component.

Illocution and cognition : the case of apologies

This paper analyzes the speech act of apologizing from the point of view of the conceptual tools provided by Cognitive Linguistics. Speech acts have not received a lot of attention in this approach to language. Notable exceptions are Panther and Thornburg (1998), Pérez (1997, 2001), Sweetser (2000), and Pérez and Ruiz de Mendoza (2002). While Sweetser's study focuses on the performative value of metaphor, the studies carried out by Panther and Thornburg, on the one hand, and Pérez and Ruiz de Mendoza, on the other, share a commitment with reinterpreting traditional speech act theory in terms of cognitive modeling. Thus, Panther and Thornburg (1998) claim that illocutionary force is a form of inferential activity grounded in metonymy. Pérez and Ruiz de Mendoza (2002), in turn, have developed a more sophisticated theory according to which the illocutionary value of an utterance is a function of a number of cognitive models which interact in different but predictable ways. Pérez and Ruiz de Mendoza have thus analyzed the interpretation of directive speech acts as a combination of metaphorical and metonymic operations performed on propositional cognitive models of interpersonal nature, in connection to pre-conceptual sensorimotor and spatial representations, such as the FORCE image-schema. Here we extend Pérez and Ruiz de Mendoza's account to the domain of expressive speech acts with an especial focus on apologies. We show that expressive speech acts are also sensitive to an analysis based on interpersonal cognitive models and image-schematic representations. In order to do so, we outline the essentials of an illocutionary scenario for apologies and explore the way it is exploited by means of other cognitive operations, especially metonymy. We additionally show that Pérez and Ruiz de Mendoza's cost-benefit idealized cognitive model, which these authors have postulated for directive and commissive speech acts, also lies at the base of our understanding of expressive categories.