Speaker Representation Research Articles

THE (UN)INTENDED CONSEQUENCES OF BILINGUAL EMPLOYMENT POLICIES

AbstractRecent immigration and migration patterns have altered the ethnoracial composition of Alameda County, California. Sociopolitical leaders have struggled to adjust to these changes. In an effort to facilitate limited English speakers’ access to critical municipal services, Oakland—the largest municipal in Alameda County—passed an Equal Access to Services Ordinance on May 8, 2001, which is a groundbreaking language access legislation for the City of Oakland’s public administration. Using data from the 2000 Census and the 2005–2011 American Community Survey, this study examines the impact of bilingual employment policies on the ethnoracial segmentation of Alameda County workers. Logistic regression reveals that bilingual employment policies have reorganized both targeted (i.e., public contact) and non-targeted occupations within the local government public administration sector. Specifically, Spanish/Chinese bilingual speakers made gains in the public administration sector (the intended effects), while Black monolingual English speakers experienced losses (the unintended effects). The representation of Black monolingual English speakers in public contact jobs within the local government public administration sector declined by as much as 18 percentage points after the implementation of the nation’s first municipal-level bilingual employment policy. The impact of bilingual employment policies on the East Bay’s Black/Brown relations and African American’s hold on low-skilled jobs in service industries is examined.

Dynamic representation of acoustic features in neural responses to continuous speech

Humans are unique in their ability to communicate using spoken language. However, it remains unclear how distinct acoustic features of speech sounds are represented in the auditory pathway over time. In this study, we applied a novel analysis technique to electroencephalography (EEG) signals as subjects listened to continuous speech and characterize the neural representation of acoustic features and the progression of responses over time. We averaged the time-aligned neural responses to phoneme instances and calculated a phoneme-related potential (PRP). We show that phonemes in continuous speech evoke multiple observable responses which are clearly separated in time. These recurrent responses have different scalp distributions, and occur as early as 50 ms, and as late as 400 ms after the phoneme onset. We show that the responses explicitly represent the acoustic distinctions of phonemes, and that linguistic and non-linguistic information appear at different time intervals. Finally, we show a joint encoding of phonetic and speaker information, where the neural representation of speakers is dependent on phoneme category. This study provides evidence for a dynamic neural transformation of low-level speech features and form an empirical framework to study the representational changes in learning, attention, and speech disorders.

Review: Austin, Jennifer, Blume, María, & Sánchez, Liliana (2015). Bilingualism in the Spanish-speaking world: Linguistic and cognitive perspectives. Cambridge, United Kingdom: Cambridge University Press.

The book reviewed here may serve as an excellent resource for advanced undergraduate students of linguistics, as well as graduates and scholars in the field and related disciplines. It comprises five chapters that explore, through three Spanish bilingual communities, the following topics: basic concepts in bilingualism, such as code-switching and language transfer; a brief historical review of Spanish bilingualism; proposed definitions of bilingualism and the different ways in which bilinguals may be classified according to factors such as language competence and age of acquisition; the effect of bilingualism on the brains and minds of bilingual speakers, such as increased cortical activation and enhanced literacy; and, finally, the development and outcomes of bilingualism on the linguistic representations of bilingual speakers. The authors cover a breadth of research, rendering it accessible by succinctly reviewing findings, and further identify areas for future research.

Representation of the Characters in the Claimed English as an International Language-Targeted Coursebooks

<p><em>The cultural and linguistic hegemony of the native speakers of English over the non-native speakers in the process of language learning and teaching has paved the way for the stereotypical and biased representations of the non-native speakers of English in majority of the English Language Teaching (ELT) coursebooks. Actually, this essentialist approach in the process of materials development is likely to result in reductionist overgeneralization and otherization of foreign societies (Holliday, 1994). However, in recent years, with the advent of English as a International Language (EIL), the issue of native speakerism, the ownership of English, and consequently the cultural content of ELT coursebooks have been the subject of debates. Despite the dominance of native speaker varieties of English in ELT coursebooks, there has been a growing awareness among publishers over the past years and accordingly some EIL-based coursebooks targeted specifically at EIL learners have been published. In this study, a sample group of such coursebooks was subject to close scrutiny. In so doing, an attempt was made to examine the representation of the characters in the claimed EIL-targeted coursebooks. According to the results of the study, despite the claim to be based on EIL, the biased representation of the non-native speakers of English is observed throughout the entire series of the analyzed coursebooks and they superficially surface a stereotypical association of culture and location/country.</em><em></em></p>

Whose Culture and How Far? Culture Presentation in Current Business English Textbook Series

International business communication is associated with cultural norms of both native and non-native business counterparts. It seems that the study of culture presentation from the perspective of English as a lingua franca in business English textbooks has been rather underexplored. To this end, cultural contents in two internationally distributed business English book series published by leading publishers were examined. It was found that the selected textbooks in the series were in favor of a representation of native speakers in global business encounters, the inner-circle culture of native speakers was the dominant culture content, and the level of cultural presentation tended to remain in the knowledge-oriented rather than communication-based level. This leads to an argument in favor of critical and reflective culture presentations in business materials.

Factor analysis method for text-independent speaker identification

Factor analysis method offers state-of-the-art performance in speaker identification during the paper. The compact representations of speakers named i-vectors are extracted from the utterances in a new low dimensional speaker- and channel-dependent space, referred to a total variability space. LBG algorithm is combined with fuzzy theory in the initialization of speaker models，which helps improve the recognition rate of the system. Channel compensation techniques, such as Linear Discriminate Analysis (LDA), Principal Component Analysis (PCA), Nuisance Attribute Projection (NAP) and Within-class Covariance Normalization (WCCN) are compared during the experiment. It can be seen that LDA followed by WCCN achieves satisfying performance. In addition, experiments contrast several identification methods. One is through Support-Vector-Machine (SVM), another one directly uses the cosine distance similarity (CDS) as the final decision score, logarithmic likelihood and vector quantization are used to compare to above two methods. It demonstrates that CDS combined with score normalization obtains better result. The testing of mobile phone database shows the robustness of the system in complex channel environment. The graphical user interface of training and testing module is simulated on MATLAB in the end of the paper.

Multiple Grammars and Second Language Representation

This paper presents an extension of the Multiple Grammars Theory (Roeper, 1999) to provide a formal mechanism that can serve as a generative-based alternative to current descriptive models of interlanguage. The theory extends historical work by Kroch and Taylor (1997), and has been taken into a computational direction by Yang (2003). The proposal is based on the idea that any human grammar readily accommodates sets of rules in sub-grammars that can seem (apparently) contradictory. We discuss the rationale behind this proposal and establish a dialogue with recent research in SLA, multilingualism, L3 acquisition, and L2 processing. We compare the Multiple Grammars explanation to optionality in L2 to other current proposals, and provide experimental results that can demonstrate the existence of active sub-grammars in the linguistic representation of L2 speakers.

In faint praise of folly

This study critically examines the widespread practice of comparing the linguistic processes and representations of non-native speakers with those of natives. We argue that, in some respects, the method yields benefits, while in others it does not serve the interests of research into the nature of second language acquisition and bilingualism. We go on to consider certain analytical approaches that skirt the hazards of the method. The potential payoffs of native/non-native comparisons are illustrated in a priming study of monolingual and bilingual processing of ambiguity in complex French syntax (Gertken 2013).

Is ELF more effective and fair than translation? An evaluation of the EU's multilingual regime

The management of multilingualism in the European Union (EU) has become an increasingly debated issue. Some argue that ‘English as a lingua franca’ (ELF) should help solve many of the attendant challenges. The claim is sometimes made that ELF, as a multilingual way of using English, is no longer dependent upon the practices and representations of native speakers and therefore poses no threat to linguistic diversity or to fairness. In this article, we question the relevance of the concept of ELF from a language policy perspective, showing that the socioeconomic implications of using ELF are no different from those of using English only. We examine the costs associated with alternative language regimes for the EU, showing that a reduction in the direct costs of European multilingualism through the imposition of a single language merely shifts costs towards citizens whose mother tongue is not English, and that re‐labelling it as ‘ELF’ makes no difference. The use of translation and interpreting, though not free, remains more effective (and at a reasonable cost) than a monolingual regime based on English alone; it also more fair than a monolingual regime which unavoidably privileges native speakers.

The Female Voice in Italian Renaissance Dialogue

A remarkable novelty within the early modern Italian tradition of literary dialogue was its inclusion in significant numbers of women among the interlocutors portrayed. Read with due sensitivity to genre conventions, these dialogues can offer insights into gendered speech decorum that interestingly nuance the evidence offered by prescriptive texts such as conduct manuals. The essay offers an overview of early modern Italian dialogues featuring female speakers, encompassing a chronological account of their development, and discussions of their production contexts and thematics, as well as an analysis of their gender dynamics.

Invisibility and Ownership of Language: Problems of Representation in Russian Language Textbooks

Using the 2 theoretical lenses of representation (Barthes, 1977) and ownership of language (Higgins, 2003), this article offers a critical analysis of representations of Russian speakers in 9 widely used Russian language textbooks aimed at university‐level learners. Particular attention is paid to representations of Russian speakers other than white Orthodox Christian ethnic Russians. Findings indicate that there are virtually no representations of such “other” speakers in the materials under investigation. This absence is framed as an erasure (hooks, 1992), a violent removal that significantly misrepresents the actual diversity of speakers of Russian and restricts putative ownership of the Russian language. The authors argue that not only is this inaccurate portrayal pedagogically problematic; it is also politically so, as it coincides with a rising tide of ethnic Russian nationalism in the Russian Federation. The article ends with a call for more accurate portrayals of the actual diversity of users of Russian.

Front-End Factor Analysis for Speaker Verification

This paper presents an extension of our previous work which proposes a new speaker representation for speaker verification. In this modeling, a new low-dimensional speaker- and channel-dependent space is defined using a simple factor analysis. This space is named the total variability space because it models both speaker and channel variabilities. Two speaker verification systems are proposed which use this new representation. The first system is a support vector machine-based system that uses the cosine kernel to estimate the similarity between the input data. The second system directly uses the cosine similarity as the final decision score. We tested three channel compensation techniques in the total variability space, which are within-class covariance normalization (WCCN), linear discriminate analysis (LDA), and nuisance attribute projection (NAP). We found that the best results are obtained when LDA is followed by WCCN. We achieved an equal error rate (EER) of 1.12% and MinDCF of 0.0094 using the cosine distance scoring on the male English trials of the core condition of the NIST 2008 Speaker Recognition Evaluation dataset. We also obtained 4% absolute EER improvement for both-gender trials on the 10 s-10 s condition compared to the classical joint factor analysis scoring.

Chastity Warrants for Women Public Speakers in Nineteenth-Century American Fiction

Accusations of sexual impropriety have been used against women public speakers at least since the Renaissance, and nineteenth-century America was no exception. In constructing public personae that worked with prevailing gender ideologies, women tried to preserve the appearance of sexual purity. This concern for chastity carried over into fictional representations of women public speakers. While some authors depicted such figures negatively, the three examined here—Elizabeth Stuart Phelps, Louisa May Alcott, and Frances E. W. Harper—all defended the woman public speaker by providing warrants within the narrative structure for her chastity and by giving her a public mission that was appropriately feminine according to nineteenth-century gender ideology. These women authors also provided a utopian moment in their narratives in which the social benefits of allowing their protagonists to speak in public are dramatized. Studying the literary representations of women speakers, in any era, can help to illuminate the cultural milieu in which such women made their way.

The Representation of Non-native Speakers in Japanese Textbook Dialogues A Case Study of Genki

The Representation of Non-native Speakers in Japanese Textbook Dialogues A Case Study of Genki

A real-time trained system for robust speaker verification using relative space of anchor models

A real-time trained system for robust speaker verification is proposed. This system was developed using a relative space of reference speakers, also referred to as anchor models. The real-time training aspect of the system is based on this relative space’s intriguing features and properties. The relative space concept uses relative speaker representation rather than an absolute representation, by comparing the speaker to a set of well-trained reference speakers. The advantage of this approach is that instead of estimating numerous parameters of an absolute model for a speaker, only a few parameters of a model relative to a number of anchor models are estimated. In order to optimize the performance of the proposed system, several techniques were assessed for possible implementation in various blocks of the system. As a result, the best performance was achieved where normalized vector’s mutual angle with the Minimum normalization method was applied to speaker verification in conjunction with an orthogonal relative space of virtual reference speakers. In this case, an Equal Error Rate (EER) of 0.12% on 400 test samples of 100 speakers was obtained. In addition to assessment under normal conditions, the developed speaker verification system was also evaluated under abnormal conditions where noisy or telephonic speech sequence contamination was present. Experiments conducted in this case demonstrated that, in most cases, this system outperforms absolute space based systems even with shortened training speech sequences. Another major contribution of this research is the development of a more complex speaker verification system capable of tackling abnormal conditions more effectively. In this case, other interesting features of the relative space approach were employed. For this purpose, a novel enrichment method was developed to construct a relative space of anchor models trained to tackle noise. The results of the experiments conducted in this part of the research demonstrated an excellent ability of this approach to tackle abnormal conditions. Compared to absolute space based system, applying this method in relative space led to lower error rates of speaker verification in all cases even with low SNR values.

Narration and Vividness as Measures of Event-Specificity in Autobiographical Memory

The event specificity of autobiographical memories refers to the degree to which retold memories include specific details about a unique personal experience from a variety of representational systems supported by different brain areas. This article proposes 2 text measures as indicators of event specificity: (a) a measure of temporal sequence in narration (Labov & Waletzky, 1967; Nelson & Horowitz, 2001) and (b) referential activity (RA; Bucci, 1997). Temporal sequence in narration measures the degree to which a personal memory includes clauses spoken in the same order as the speaker's representation of events. RA is text vividness, or the extent to which words clearly refer to a speaker's detailed cognitive, sensory, and emotional experiences. This article hypothesizes that temporal sequences and RA are both signs of speaker activation of event-specific memories and should be positively correlated. Using 55 transcripts from high school students interviewed about their “most stressful time,” this article analyzes the first 250 words of each memory for narrative temporal sequences and for RA. Number of temporal sequences and RA scores were positively associated (Spearman's r = .69, p < .001), consistent with the hypothesis.

BILINGUAL MINDS: EMOTIONAL EXPERIENCE, EXPRESSION AND REPRESENTATION, Aneta Pavlenko (Ed.)

BILINGUAL MINDS: EMOTIONAL EXPERIENCE, EXPRESSION AND REPRESENTATION. Aneta Pavlenko (Ed.). Clevedon, UK: Multilingual Matters, 2006. Pp. xvi + 324. $54.95 paper.Pioneering research by Pavlenko (2002) has attracted the attention of many researchers to the emotions of bilinguals and multilinguals. The present collection of chapters reports a number of recent studies in this new research area. The 11 contributions, written by established academics, cover various issues in emotional experience, emotional expression, and emotional representation of bilingual and multilingual speakers.

Introduction: "Confirm my voice": "My sisters," Poetic Audiences, and the Published Voices of EBB

Introduction "Confirm my voice": "My sisters," Poetic Audiences, and the Published Voices of EBB Marjorie Stone Beverly Taylor In a pocket notebook including works dating from the 1842-44 period, the poet who used the name Elizabeth Barrett Barrett before her marriage and Elizabeth Barrett Browning afterwards, or "EBB" for short throughout her life,1 drafted an untitled, unfinished poem beginning "My sisters! Daughters of this Fatherland / Which we call England!"2 In this fragment, published here for the first time (see below), she calls upon the women of England: "Give me your ear & heart—Grant me yr voice / Do confirm my voice—lest it speak in vain." Fifty years after her birth in 1806, there seemed little reason to believe that Elizabeth Barrett Browning had spoken "in vain." By 1856, hosts of readers—male as well as female, in England and well beyond its borders—"confirmed" her voice with widespread reviews, frequent tributes, strong disagreements (an indirect form of confirmation), and solid sales. Her 1844 Poems (published in the United States as A Drama of Exile: And Other Poems), combined with works such as "The Runaway Slave at Pilgrim's Point," published in the 1848 issue of the Boston anti-slavery annual The Liberty Bell, had established her trans-Atlantic reputation. By mid century, she stood with Tennyson among the first rank of English poets, celebrated not only by the public, but also by other writers and artists. She was the only woman writer included in the list of "Immortals" drawn up by the zealous young Pre-Raphaelite Brotherhood in 1848.3 She was also named by the Athenaeum as a suitable contender for the position of Poet Laureate on Wordsworth's death in 1850, the year in which an expanded new edition of her Poems appeared.4 Casa Guidi Windows (1851), on the movement for Italian unification and nationhood, generated mixed reviews in England, given its controversial call for British intervention in continental politics. In Italy, however, in an 1852 address to the Piedmont Chamber of Deputies, it was cited by the Italian writer and politician, Massimo D'Azeglio, then Prime Minister of the Italian state, and praised even by some of the Italian patriots whom the poem [End Page 391] critiqued.5 Aurora Leigh (1856), the first extended portrait of the woman poet in English literature, further enhanced EBB's international reputation even as it generated animated debates, going into more than twenty editions by century's end. Meanwhile, notwithstanding the backlash in England against Poems before Congress (1860), the poet's second book calling for British intervention in Italian liberation, her collected poems were repeatedly reprinted, or issued in new editions, along with multiple selected editions. In the second half of the nineteenth century, her work was discussed in France, especially popular in North America, and translated into French and Italian—even, in the case of "The Cry of the Children," into Russian.6 While hosts of nineteenth readers in different nations and languages may have "confirmed" EBB's voice with ear, heart, writing, and speech acts of their own, their divergent responses to a body of work that is extraordinarily wide-ranging in its representation of speakers, deployment of poetic forms, and variable subject matter suggests how complex and multi-faceted an issue "voice" can be in her poetry. To accommodate the amplitude of her poetry and multiplicity of its effects—its intertextual engagements with precursors, its refraction through varying generic registers, and the "Uproar in the echo"7 that it excited through its utterance—we might do better to speak of the published "voices" of EBB's poetry. Our attachment to the concept of a holistic voice in reading poetry—a voice that connects us to the poet herself or himself through a subtle medium as intimate as breathing—-speaks to our nostalgia for the integrated unitary self as an ultimate origin of utterance. Yet as Margaret Linley, Eric Griffiths, and Yopie Prins, among...

Speaker recognition by location in the space of reference speakers

Speaker representation by location in a reference space is a new technique of speaker recognition and adaptation. It consists in representing a speaker relatively rather than absolutely, by comparing him to a set of well-trained speakers. The main motivation is to obtain a compact modeling of every speaker, which gives similar performances to those of the state of the art GMM-UBM. Thus, instead of estimating numerous parameters of an absolute model of the speaker, only a few parameters of a model relatively to other speaker models called reference speakers are estimated. In this study, several points are addressed that are related to the concept of relative location in speaker recognition. Firstly, the reference speaker space is built. Then the appropriate metrics in this space are investigated in order to perform speaker recognition in a geometrical approach. Finally, a statistical approach for speaker location is used to eliminate the weaknesses of the geometrical approach. In-depth evaluations on a telephone database show that the concept of relative location is a promising technique for speaker verification. Therefore, it can be concluded that the most important motivation for using anchor models is their computational efficiency for indexing tasks.

On the Determination of Optimal Model Order for GMM-Based Text-Independent Speaker Identification

Gaussian mixture models (GMMs) are recently employed to provide a robust technique for speaker identification. The determination of the appropriate number of Gaussian components in amodel for adequate speaker representation is a crucial but difficult problem. This number is in fact speaker dependent. Therefore, assuming a fixed number of Gaussian components for all speakers is not justified. In this paper, we develop a procedure for roughly estimating the maximum possible model order above which the estimation of model parameters becomes unreliable. In addition, a theoretical measure, namely, a goodness of fit (GOF) measure is derived and utilized in estimating the number of Gaussian components needed to characterize different speakers. The estimation is carried out by exploiting the distribution of the training data for each speaker. Experimental results indicate that the proposed technique provides results comparable to other well-known model selection criteria like the minimum description length (MDL) and the Akaike information criterion (AIC).