Search

With the enormous increase in accessibility of high-speed internet, the number of social media users is increasing rapidly. Due to a lack of proper regulations and ethics, social media platforms are often contaminated by posts and comments containing abusive language and offensive remarks toward individuals, groups, races, religions, and communities. A single remark often triggers a huge chain of reactions with similar abusiveness, or even more. To prevent such occurrences, there is a need for automated systems that can detect abusive texts and hate speeches and remove them immediately. However, most existing research works are limited only to globally popular languages like English. Since India is a nation of many diverse languages and multiple religions, nowadays abusive posts and remarks in Indian languages (monolingual or code-mixed form) are not infrequent on social media platforms. Although resources such as hate speech lexicon and annotated datasets are limited for Indian languages, most research works on hate speech detection in such languages used traditional machine learning and deep learning methods for this task. However, multilingualism and code-mixing make hate speech detection in Indian languages more challenging. Given these facts, this paper mainly focuses on reviewing the latest impactful research works on hate speech detection in Indian languages. In this paper, we have analyzed and compared the latest research works on hate speech detection in Indian languages in terms of various aspects—datasets used, feature extraction and classification methods applied, and the results achieved. [ABSTRACT FROM AUTHOR]

This research underlines the communication patterns of Indonesian diaspora women in their mixed culture families and how they use these patterns in Australia. There are some differences in interpreting meanings because of the differences in language and culture. Thus, researchers are interested in examining communication patterns experienced by these diaspora women who live in Australia. The research was conducted through a qualitative method with a phenomenological approach. Data collection techniques used were in-depth interviews with 13 Indonesian women who had been living in Australia for more than 5 years and observation and documentation studies. The study showed that the communication patterns of Indonesian diaspora women living within their mixed culture families consist mainly of verbal communication classified into five categories: (1) pure English, (2) mixed language, (3) unstructured grammar English, (4) compliments and appreciation, and (5) obscene words or swearing. [ABSTRACT FROM AUTHOR]

The textbook account of the history of English suggests that Middle English favoured lexical borrowing and native compounding declined. Still, some types of compounds carried on being productive and new types were developed in Middle English. The non-negligible presence of complex lexical units—encompassing both compounds and other multi-word expressions (see Schlücker 2019)—in late medieval English texts whose base language is Medieval Latin is worth tracing: What kinds of complex lexical units are present in those texts and why are they expressed in the vernacular instead of using Latin equivalents (if any)? This article gives a sketch of the background to the historical development of compounding, putting it alongside (rather than in isolation from) the languages and textual traditions with which English had contact, and provides examples of such compound formations from a case study, the Durham Account Rolls. [ABSTRACT FROM AUTHOR]

Speech emotion recognition continues to attract a lot of research especially under mixed-language scenarios. Here, we show that emotion is language dependent and that enhanced emotion recognition systems can be built when the language is known. We propose a two-stage emotion recognition system that starts by identifying the language, followed by a dedicated language-dependent recognition system for identifying the type of emotion. The system is able to recognize accurately the four main types of emotion, namely neutral, happy, angry, and sad. These types of emotion states are widely used in practical setups. To keep the computation complexity low, we identify the language using a feature vector consisting of energies from a basic wavelet decomposition. A hidden Markov model (HMM) is then used to track the changes of this vector to identify the language, achieving recognition accuracy close to 100%. Once the language is identified, a set of speech processing features including pitch and MFCCs are used with a neural network (NN) architecture to identify the emotion type. The results show that that identifying the language first can substantially improve the overall accuracy in identifying emotions. The overall accuracy achieved with the proposed system reached more than 93%. To test the robustness of the proposed methodology, we also used a Gaussian mixture model (GMM) for both language identification and emotion recognition. Our proposed HMM-NN approach showed a better performance than the GMM-based approach. More importantly, we tested the proposed algorithm with 6 emotions which are showed that the overall accuracy continues to be excellent, while the performance of the GMM-based approach deteriorates substantially. It is worth noting that the performance we achieved is close to the one attained for single language emotion recognition systems and outperforms by far recognition systems without language identification (around 60%). The work shows the strong correlation between language and type of emotion, and can further be extended to other scenarios including gender-based, facial expression-based, and age-based emotion recognition. [ABSTRACT FROM AUTHOR]

In this article, I demonstrate how hybrid language practices allow for children to make use of their full linguistic repertoire in negotiating the social and communicative demands of the school environment (Axelrod in Early Child Educ J, p 1-8, 2014). I argue that the unusual case of classe LSF and École Maternelle Gabriel Sajus directly challenges mainstream progressive assumptions of inclusion that effectively exclude deaf children from their deaf peers. Current so-called progressive models of inclusion often isolate deaf learners in mainstream classroom settings with non-deaf classmates and adults. Classe LSF and École Maternelle Gabriel Sajus offers insights for rethinking inclusion in early childhood classrooms in the ways that it provokes us to consider the inclusive potential of hybrid linguistic spaces for bilingual learners. [ABSTRACT FROM AUTHOR]

Copyright of Russian Linguistics is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

We define the notion of controlled hybrid language that allows information share and interaction between a controlled natural language (specified by a context-free grammar) and a controlled visual language (specified by a Symbol-Relation grammar). We present the controlled hybrid language INAUT, used to represent nautical charts of the French Naval and Hydrographic Service (SHOM) and their companion texts ( Instructions nautiques). [ABSTRACT FROM AUTHOR]

Substantial amounts of resources are usually required to robustly develop a language model for an open vocabulary speech recognition system as out-of-vocabulary (OOV) words can hurt recognition accuracy. In this work, we applied a hybrid lexicon of word and sub-word units to resolve the problem of OOV words in a resource-efficient way. As sub-lexical units can be combined to form new words, a compact set of hybrid vocabulary can be used while still maintaining a low OOV rate. For Thai, a syllable-based unit called pseudo-morpheme (PM) was chosen as a sub-word unit. To also benefit from different levels of linguistic information embedded in different input types, a hybrid recurrent neural network language model (RNNLM) framework is proposed. An RNNLM can model not only information from multiple-type input units through a hybrid input vector of words and PMs, but can also capture long context history through recurrent connections. Several hybrid input representations were also explored to optimize both recognition accuracy and computational time. The hybrid LM has shown to be both resource-efficient and well-performed on two Thai LVCSR tasks: broadcast news transcription and speech-to-speech translation. The proposed hybrid lexicon can constitute an open vocabulary for Thai LVCSR as it can greatly reduce the OOV rate to less than 1 % while using only 42 % of the vocabulary size of the word-based lexicon. In terms of recognition performance, the best proposed hybrid RNNLM, which uses a mixed word-PM input, obtained 1.54 % relative WER reduction when compared with a conventional word-based RNNLM. In terms of computational time, the best hybrid RNNLM has the lowest training and decoding time among all RNNLMs including the word-based RNNLM. The overall relative reduction on WER of the proposed hybrid RNNLM over a traditional n-gram model is 6.91 %. [ABSTRACT FROM AUTHOR]

For any given mixed-language text, a multilingual synthesizer synthesizes speech that is intelligible to human listener. However, as speech data are usually collected from native speakers to avoid foreign accent, synthesized speech shows speaker switching at language switching points. To overcome this, the multilingual speech corpus can be converted to a polyglot speech corpus using cross-lingual voice conversion, and a polyglot synthesizer can be developed. Cross-lingual voice conversion is a technique to produce utterances in target speaker's voice from source speaker's utterance irrespective of the language and text spoken by the source and the target speakers. Conventional voice conversion technique based on GMM tokenization suffer from degradation in speech quality as the spectrum is oversmoothed due to statistical averaging. The current work focuses on alleviating the oversmoothing effect in GMM-based voice conversion technique, using (source) language-specific mixture weights in a multi-level GMM followed by selective pole focusing in the unvoiced speech segments. The continuity between the frames of the converted speech is ensured by performing fifth-order mean filtering in the cepstral domain. For the current work, cross-lingual voice conversion is performed for four regional Indian languages and a foreign language namely, Tamil, Telugu, Malayalam, Hindi, and Indian English. The performance of the system is evaluated subjectively using ABX listening test for speaker identity and using mean opinion score for quality. Experimental results demonstrate that the proposed method effectively improves the quality and intelligibility mitigating the oversmoothing effect in the voice-converted speech. A hidden Markov model-based polyglot text-to-speech system is also developed, using this converted speech corpus, to further make the system suitable for unrestricted vocabulary. [ABSTRACT FROM AUTHOR]

Hybrid languages are introduced in order to evaluate the strength of 'minimal' mereologies with relatively strong frame definability properties. Appealing to a robust form of nominalism, I claim that one investigated language $\mathcal {H}_{\textsf {m}}$ is maximally acceptable for nominalistic mereology. In an extension $\mathcal {H}_{\textsf {gem}}$ of $\mathcal {H}_{\textsf {m}}$, a modal analog for the classical systems of Leonard and Goodman (J Symb Log 5:45-55, 1940) and Leśniewski (1916) is introduced and shown to be complete with respect to 0-deleted Boolean algebras. We characterize the formulas of first-order logic invariant for $\mathcal {H}_{\textsf {gem}}$-bisimulations. [ABSTRACT FROM AUTHOR]

This paper presents the layout description script (LDS), which is a domain specific language intended to code layout templates to be used for layout-aware circuit synthesis. LDS supports both sequential and constraint programming and is suitable for both manual coding and automatic code generation. LDS is compared with previous approaches related to layout description. Code samples are given for alignment, abutment, symmetry, and similar constraints. Also, implementation of the LDS compiler is discussed and a methodology for handling complex constraints is presented. Due to its support for constraint programming, it is possible to constrain topological representations and even combine them. It is also possible to combine and constrain placement and routing in an LDS template. Finally, a capture tool has been implemented. This tool is designed to extract a template from an expert-drawn layout. Capture converts a data structure extracted through a guided user interface into a template. This tool highlights the compatibility of LDS with electronic design automation. [ABSTRACT FROM AUTHOR]

Uspanteko (Guatemala; ∼2000 speakers) is an endangered K'ichean-branch Mayan language. It is unique among the K'ichean languages in having innovated a system of contrastive pitch accent, which operates alongside a separate system of non-contrastive stress. The prosody of Uspanteko is of general typological interest, given the relative scarcity of 'mixed' languages employing both stress and lexical pitch. Drawing from a descriptive grammar and from our own fieldwork, we also document some intricate interactions between pitch accent and other aspects of the phonology (stress placement, vowel length, vowel quality, and two deletion processes). While pitch accent is closely tied to morphology, the location of lexical tone is entirely a matter of surface phonology. We propose that the position of pitch accent and stress is determined by three factors: (i) feet are always right-aligned, and preferably iambic; (ii) pitch accent must fall on a stressed syllable; and (iii) pitch accent cannot fall on a final mora. These assumptions derive default final stress, as well as a regular pattern of tone-triggered stress shift. Interactions between prosody and segmental phonotactics are attributed to further constraints on footing. Surprisingly, we find robust evidence for foot structure in Uspanteko, even though these patterns could easily be described in non-metrical terms. Interactions between tone and vowel length also provide evidence for lexical strata within the Uspanteko vocabulary. [ABSTRACT FROM AUTHOR]

The strong historical link between the Canary Islands and the Americas has facilitated the existence of a tight lexical relationship between both territories. This connection began right from the period of their conquest and colonization. So, the creation of the new lexical repertoires in the Canaries and America shares the same processes, timing and features. But this likeness, often studied and justified, could not hide the peculiarities of the Canarian lexicon that follow from the historical and social facts of the Canarian Europeanization process. These singularities appear in different jargons, but perhaps it is specially evident in the case of animal names: the lack of terms derived from the language of the guanches, the scarce presence of words coming from the west of the Spanish mainland, contrary to what happens in other repertoires, and the degree of wealth and the spread of literacy, are the most important characteristics of this lexical field. [ABSTRACT FROM AUTHOR]

Language acquisition research in autism has traditionally focused on high-level pragmatic deficits. Few studies have examined grammatical abilities in autism, with mixed findings. The present study addresses this gap in the literature by providing a detailed investigation of syntactic and higher-level discourse abilities in verbal children with autism, age 5 years. Findings indicate clear language difficulties that go beyond what would be expected based on developmental level; specifically, syntactic delays, impairments in discourse management and increased production of non-meaningful words (jargon). The present study indicates a highly specific pattern of language impairments, and importantly, syntactic delays, in a group of children with autism carefully matched on lexical level and non-verbal mental age with children with developmental delays and typical development. [ABSTRACT FROM AUTHOR]

Following a brief contextualization of bilingual Cuban-American literature in relation to both Latino writing and Cuban exile fiction, this article moves on to consider the meaning and placement of both the Spanish and English languages in Gustavo Pérez Firmat's poetry collection, "Bilingual Blues." It then turns to a close analysis of two of the mixed language poems of this text, namely "Bilingual Blues" and "Turning the Times Tables." These are discussed in terms of suggested cultural affiliations and identifications, as indicated in the poet's use of not only languages but also imagery and intertextual reference. Such factors are also connected with certain theoretical debates regarding migration and identity in this article. [ABSTRACT FROM AUTHOR]

Statistical methods to extract translational equivalents from non-parallel corpora hold the promise of ensuring the required coverage and domain customisation of lexicons as well as accelerating their compilation and maintenance. A challenge for these methods are rare, less common words and expressions, which often have low corpus frequencies. However, it is rare words such as newly introduced terminology and named entities that present the main interest for practical lexical acquisition. In this article, we study possibilities of improving the extraction of low-frequency equivalents from bilingual comparable corpora. Our work is carried out in the general framework which discovers equivalences between words of different languages using similarities between their occurrence patterns found in respective monolingual corpora. We develop a method that aims to compensate for insufficient amounts of corpus evidence on rare words: prior to measuring cross-language similarities, the method uses same-language corpus data to model co-occurrence vectors of rare words by predicting their unseen co-occurrences and smoothing rare, unreliable ones. Our experimental evaluation demonstrates that the proposed method delivers a consistent and significant improvement on the conventional approach to this task. [ABSTRACT FROM AUTHOR]

Code-switching is very common among bilingual speakers. Spoken queries by these speakers are typically in mixed language. In this paper, we propose an unsupervised method for mixed-language query understanding, using only a monolingual corpus and a bilingual dictionary. Secondary-language words mixed in a primary-language query are translated into words in the primary language. We found that using a single disambiguation feature for translation is more effective than using multiple features, provided this feature is based on the most salient seed-word, chosen automatically by confidence scoring. We propose and compare four types of disambiguation features that are based on context seed-words. A baseline method uses the nearest neighboring seed-word as disambiguation feature. Multiple-context seed-word voting is also proposed in order to enlarge the context window. On the other hand, merely using the inverse-distance as weights on context words degrades the performance as it runs counter to the potential underlying syntactic relations between words. Our final proposal is a solution that uses multiple-context seed-words and the translation candidates of all mixed language words to select a single most salient seed-word for translation disambiguation. The translation disambiguation accuracy for this feature is at 83.7% for all words in the ATIS spontaneous speech query database, and 66.7% for content words. [ABSTRACT FROM AUTHOR]

Explores the degree to which sentence context effects operate at a lexical or conceptual level. Processing of mixed-language sentences by fluent Spanish-English bilinguals; Sentence context effects as influenced by both semantic/conceptual and lexical information; Facilitation in processing a contextually constrained target word.

A letter to the editor is presented in response to an article by Trevor Quirk about improving communication in science and the need for science writers to not shun away from scientific jargons.