Search

This paper describes an application based on internet of things (IoT) and gesture recognition through the MediaPipe library to communicate people with hearing and speech disabilities who use Mexican sign language (MSL) with people who do not have a disability. The system is made up of two modules, on the one hand, there is an Apple watch with the ability to obtain text and voice inputs in Spanish to stand for the translation into sign language through video and images. The system can set up a connection with an application in charge of performing sign recognition by capturing images from a camera, the IoT user can connect so that communication between both modules is carried out bidirectionally through Firebase database. During the development of the experimental tests, the visual recognition module was able to recognize signs belonging to the Spanish alphabet, digits, and greetings, with a 96% of precision, while the IoT device correctly displayed voice translations and text to symbology compared to the MSL. [ABSTRACT FROM AUTHOR]

Copyright of Ordnance Industry Automation is the property of Editorial Board for Ordnance Industry Automation 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.)

Yarrie Lingo is the local name for the English-lexified creole language spoken in the Aboriginal community of Yarrabah in far north-eastern Queensland, Australia. This creole has only recently been gaining recognition but it is the main language of everyday interactions in Yarrabah. This study describes the recognitional trajectory of Yarrie Lingo and what has fostered this language awareness. With a focus on the experiences of the local Aboriginal educator, language activist and researcher, Bernadine Yeatman, we showcase a range of grounded methodologies, including devices like language posters, interviews and illustrative stories, along with maps, timelines and cartoons developed in conjunction with a linguist, Denise Angelo. This shines a light onto the practical 'how to' of language awareness for under-recognised language communities and the potential support a linguist can provide. It exemplifies a research approach which engages linguistic knowledge with local language expertise to co-design materials for community members, educators and service providers. [ABSTRACT FROM AUTHOR]

Aiming at the problem of low language recognition rate under low signal-to-noise ratio, a language recognition method based on fractional wavelet transform was proposed.Firstly, the adaptive filtering algorithm was used to filter the noise of the noisy signal, so as to reduce the influence of noise on the feature extraction and improve the processing ability of the system for non-stationary signals.Secondly, the motion of the signal on the basilar membrane of the cochlea was simulated, and then the signal was compressed by a nonlinear power function.Finally, the improved CFCC were extracted by simulating the human hearing process.Experiments show that compared with the traditional CFCC, the language recognition rate is significantly improved, and the language recognition rate is increased by 11.1% on average under the 0 dB signal-to-noise ratio, which verifies the effectiveness and robustness of the proposed algorithm.

The literature suggests ethnic diversity has a negative effect on development. Yet, we also know that government policies—e.g., recognizing multiple languages in minority-sizable areas—can attenuate these effects. In this paper, we ask: What are the socioeconomic implications of minority language recognition? We leverage a legal stipulation in Romania as a quasi-experiment: Minority languages are recognized as official in areas where the minority constitutes more than 20% of the population. We argue the recognition of minority languages builds social trust and facilitates efficiency in economic exchanges—mollifying the otherwise detrimental consequences of diversity. Using data at the municipality level, we find that in areas where only Romanian is recognized, ethnic diversity has a negative effect on development—a result consistent with the literature. This effect, however, is absent in areas where a minority language is recognized. The implications suggest that lowering the threshold for language recognition could promote even further development. [ABSTRACT FROM AUTHOR]

Copyright of Journal on Communication / Tongxin Xuebao is the property of Journal on Communications Editorial Office 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.)

Hyperdimensional computing (HDC) is an emerging computational framework inspired by the brain that operates on vectors with thousands of dimensions to emulate cognition. Unlike conventional computational frameworks that operate on numbers, HDC, like the brain, uses high-dimensional random vectors and is capable of one-shot learning. HDC is based on a well-defined set of arithmetic operations and is highly error resilient. The core operations of HDC manipulate HD vectors in bulk bit-wise fashion, offering many opportunities to leverage parallelism. Unfortunately, on conventional von Neumann architectures, the continuous movement of HD vectors among the processor and the memory can make the cognition task prohibitively slow and energy intensive. Hardware accelerators only marginally improve related metrics. In contrast, even partial implementations of an HDC framework inside memory can provide considerable performance/energy gains as demonstrated in prior work using memristors. This article presents an architecture based on racetrack memory (RTM) to conduct and accelerate the entire HDC framework within memory. The proposed solution requires minimal additional CMOS circuitry by leveraging a read operation across multiple domains in RTMs called transverse read (TR) to realize exclusive-or (XOR) and addition operations. To minimize the CMOS circuitry overhead, an RTM nanowire-based counting mechanism is proposed. Using language recognition as the example workload, the proposed RTM HDC system reduces the energy consumption by 8.6× compared to the state-of-the-art in-memory implementation. Compared to dedicated hardware design realized with an FPGA, RTM-based HDC processing demonstrates 7.8× and 5.3× improvements in the overall runtime and energy consumption, respectively. [ABSTRACT FROM AUTHOR]

This paper surveys developments in language politics and policy in Taiwan under Tsai Ing-wen's presidency (2016–present). Drawing on historical-institutionalist premises, it shows that recent language policy developments were path-dependent and built upon initiatives proposed under Chen Shui-bian's presidency (2000–2008). The paper argues that the comparative success of Tsai's initiatives owed not only to her party forming a legislative majority, but also to a broad sociopolitical consensus on transitional and historical justice, and to an incrementalist strategy that consisted in legislating on minority languages before laying out a comprehensive multilingual legal framework. Although recent language developments do fall within the purview of identity politics, these factors have enabled the Tsai administration to justify and legitimise measures towards language recognition and revitalisation as intrinsic to Taiwan's democratic consolidation, rather than as tools for identity building. [ABSTRACT FROM AUTHOR]

The article considers theoretical and practical issues of development and implementation, as well as potential development of technological solutions for voice communication of people speaking foreign languages using simultaneous machine translation, as well as potential directions of these programs for international conferences, seminars, business meetings, education in foreign universities. The article deals with the concrete experience of implementing a scientific and technological project on software development and application development carried out within the framework of the innovation project of the Russian Foundation for Innovations Development. The potential of development of this type of technologies and products in Russia in conditions of digital economy creation is also analyzed.

Deep learning methods for language recognition have achieved promising performance. However, most of the studies focus on frameworks for single types of acoustic features and single tasks. In this paper, we propose the deep joint learning strategies based on the Multi-Feature (MF) and Multi-Task (MT) models. First, we investigate the efficiency of integrating multiple acoustic features and explore two kinds of training constraints, one is introducing auxiliary classification constraints with adaptive weights for loss functions in feature encoder sub-networks, and the other option is introducing the Canonical Correlation Analysis (CCA) constraint to maximize the correlation of different feature representations. Correlated speech tasks, such as phoneme recognition, are applied as auxiliary tasks in order to learn related information to enhance the performance of language recognition. We analyze phoneme-aware information from different learning strategies, like joint learning on the frame-level, adversarial learning on the segment-level, and the combination mode. In addition, we present the Language-Phoneme embedding extraction structure to learn and extract language and phoneme embedding representations simultaneously. We demonstrate the effectiveness of the proposed approaches with experiments on the Oriental Language Recognition (OLR) data sets. Experimental results indicate that joint learning on the multi-feature and multi-task models extracts instinct feature representations for language identities and improves the performance, especially in complex challenges, such as cross-channel or open-set conditions. [ABSTRACT FROM AUTHOR]

The performance of language recognition system is mainly determined by feature extraction and model training. In this paper, a robust equalization feature for language recognition is proposed, which utilizes the common features of the speech spectrum mean vector to calculate a global mean vector. The spectrum mean vector of each segment is equalized on the global mean vector, and the equalization features are obtained. In model training, Gated Recurrent Unit (GRU) of Recurrent Neural Network (RNN) is applied to language recognition, in which GRU can reduce the amount of computation and shorten the training time. The experimental results show that the proposed method outperforms the baseline system on the NIST LRE 2007 corpus. [ABSTRACT FROM AUTHOR]

This article analyzes vector representation of phonemes as an alternative to improve a language identification system (LID). CBOW (Continuous Bag-of-Words) and Skip-gram architectures proposed by Mikolov are studied. These models allow predicting words within a context by generating n-dimensional vectors. In this work we will analyze the application of these models in smaller phonetic units or n-grams.

In order to alleviate the mismatch in model between training and testing samples caused by inter-language variations,adaptive Gaussian back-end based on LDOF criterion was proposed for language recognition.The local distance-based outlier factor (LDOF) criterion was defined to find the appropriate model parameters and dynamically select the training data subset similar to the testing samples from multiple class training sets.Then original back-end was adjusted to obtain a more matched recognition model.Experimental results on NIST LRE 2009 easily-confused language data set show that proposed method achieves an obvious performance improvement on both the equal error rate (ERR) and average decision cost function.

• A ResNet-50 with self-attention effectively learns language-dependent representations. • Convolution over the time dimension along with data-driven learned features pooling yield robust representations of languages. • The output posterior of a trained convolutional neural network can be directly used as score for verification tasks outperforming trained backends. • The combination of cross-entropy and triplet loss minimization yields discriminative representations. The problem of language identification from speech is tackled in this work. Residual convolutional neural networks are employed to this end, aiming at exploiting the ability of such architectures to take into account large contextual segments of input data. Moreover, learnable attention mechanisms are introduced on top of the convolutional stack for data-driven feature pooling across time, enabling the computation of fixed-dimension representations given varying-length speech segments as input. Training is performed using a combination of language identification and metric learning via triplet loss minimization, aimed at enforcing class separability within the embeddings space. Evaluation is performed across different conditions, such as multi-class classification, short-duration test utterances, and confusing languages, for the closed-set case, while open-set performance is evaluated with the introduction of unseen languages. At test time, end-to-end scoring along with cosine similarity and PLDA are employed, outperforming state-of-the-art benchmark methods, such as i-vectors by improving the average cost by 30% to 40% depending on the evaluation condition. [ABSTRACT FROM AUTHOR]

This paper proposes state-of the-art Automatic Speaker Recognition System (ASR) based on Bayesian Distance Learning Metric as a feature extractor. In this modeling, I explored the constraints of the distance between modified and simplified i-vector pairs by the same speaker and different speakers. An approximation of the distance metric is used as a weighted covariance matrix from the higher eigenvectors of the covariance matrix, which is used to estimate the posterior distribution of the metric distance. Given a speaker tag, I select the data pair of the different speakers with the highest cosine score to form a set of speaker constraints. This collection captures the most discriminating variability between the speakers in the training data. This Bayesian distance learning approach achieves better performance than the most advanced methods. Furthermore, this method is insensitive to normalization compared to cosine scores. This method is very effective in the case of limited training data. The modified supervised i-vector based ASR system is evaluated on the NIST SRE 2008 database. The best performance of the combined cosine score EER 1.767% obtained using LDA200 + NCA200 + LDA200, and the best performance of Bayes_dml EER 1.775% obtained using LDA200 + NCA200 + LDA100. Bayesian_dml overcomes the combined norm of cosine scores and is the best result of the short2-short3 condition report for NIST SRE 2008 data. [ABSTRACT FROM AUTHOR]

Highlights • The preprocessing algorithm removes the noise from a continuously measured dataset. • The dataset is visualized as a DNA-strain and the process recipe as a gene sequence. • The real-time integration of the algorithm in the operations assures data integrity. • The outcome is a noise-free and structured data suitable for making decisions. • A new depiction of root causes provides a fast and quantitative decision-making. Abstract The quality of data plays a crucial role in providing a reliable decision-making process when improving processes and operations under uncertainty. We present a data mining-based algorithm for robustly pre-processing the manufacturing records of biopharmaceutical batch processes. The algorithm can identify the time intervals in which the process is in commercial operation, and can characterize process failures automatically. An approximate string-matching algorithm, a decision tree classifier and a constrained clustering is applied to sequence the raw data, to classify the noise and identify each single batches; finally process failure are characterized. The algorithm was applied to the records of the process named as "cleaning- and sterilizing-in-place", which is an essential process in manufacturing environment, in a case study. The algorithm was training on state of the art manual pre-processing outcome and was applied reducing the execution time of the activity down to 11.7% while maintaining high data quality and integrity. [ABSTRACT FROM AUTHOR]

Abstract We combine Total Variability algorithm with Deep Learning theory to complete the language recognition task. The Total Variability algorithm can compensate for the influence of differences in channels and speakers among various languages, while deep learning methods have a stronger ability of nonlinear modeling compared with traditional statistical models. In this paper, I-vector feature is extracted using Total Variability algorithm, and model training is established using fully connected neural network. Meanwhile, the dropout strategy is also used to suppress overfitting. The experimental results show that the new system outperforms the baseline system on the NIST LRE 2007 corpus. [ABSTRACT FROM AUTHOR]

Across Australia, language contact has shaped the languages that many Aboriginal and Torres Strait Islander people speak today. While in some communities Traditional Languages have been spoken continuously from pre-invasion through to the present day, the language use of many First Nations Australians has shifted to "New Languages", which are contact languages such as creoles and mixed languages, or otherwise to a variety of English. Despite the extent of language contact and its effects on the day-to-day language experiences of many Indigenous Australians, there is little discourse in the public domain or in educational settings about language contact or contact languages. In public data sets they are not reliably differentiated from Traditional Languages on the one hand, nor from English(es) on the other (Angelo et al., 2019). This has profound implications for Indigenous Australians who speak these New Languages, some of which constitute the largest languages that are spoken almost exclusively by First Nations peoples today. There is no systematic policy effort underway to work with communities to describe, recognise and name these New Languages, and consequently little tailoring of services to speakers on account of these languages. This paper presents a grounded translational research methodology, Community Vernacular Language Posters (Department of Education [Qld], 2018), that has been developed for working with Aboriginal and Torres Strait Islander community members to raise awareness about their local contact language and about the ramifications for education and other service provision. The outcome of this linguistically-informed process is an accessible poster depicting local everyday ways of talking. Underpinning this document is the shared body of knowledge amongst communityparticipants that has been generated by the many consultations and inclusive discussions that have taken place due to the collaborative language research process. Often, a Community Vernacular Language Poster project has resulted in a community initiative to name their local contact language. The posters have proven a popular initiative, as they make tangible a community's speech variety, foster a sense of pride and ownership in the local vernacular and provide the community with tools for informing incomers about the local language situation. [ABSTRACT FROM AUTHOR]

For almost 30 years, digital nonlinear editing systems (DNLEs) have functioned as a practical replacement for film and videotape editing. While DNLEs have progressed in their capabilities by offering increased video resolutions and visual effects, they have not fundamentally changed their operational constructs. Editors must still choose in and outpoints of shots and then methodically edit those shots into a cohesive sequencing. Technology improvements in artificial intelligence and machine learning have the potential to profoundly impact how DNLE systems operate, and, in turn, content creation methodologies will dramatically change. This paper examines the effects of image recognition, natural speech processing, language recognition, cognitive metadata extraction, tonal analysis, and data and statistical integration on the creation of content. [ABSTRACT FROM AUTHOR]

Decision problems addressed by cellular automata have been historically expressed either as determining whether initial configurations would belong to a given language, or as classifying the initial configurations according to a property in them. Unlike traditional approaches in language recognition, classification problems have typically relied upon cyclic configurations and fully paralell (two-way) update of the cells, which render the action of the cellular automaton relatively less controllable and difficult to analyse. Although the notion of cyclic languages have been studied in the wider realm of formal languages, only recently a more systematic attempt has come into play in respect to cellular automata with fully parallel update. With the goal of contributing to this effort, we propose a unified definition of classification problem for one-dimensional, binary cellular automata, from which various known problems are couched in and novel ones are defined, and analyse the solvability of the new problems. Such a unified perspective aims at increasing existing knowledge about classification problems by cellular automata over cyclic configurations and parallel update. [ABSTRACT FROM AUTHOR]

New advances in Language Identification (LID) using Recurrent Neural Networks (RNNs) and Neural Embeddings have been proposed recently. While their application has been successfully applied at a word level, results at a phoneme level may not be as good because of the greater variability found in phoneme sequences which reduces LID accuracy. Thus, we propose to use phonetic units called “phone-grams” that implicitly include longer-context information and use them to train neural embeddings and RNN language models (RNNLMs). Neural embeddings are used in a pre-processing data phase to reduce the scattering problem produced by the high number of resulting phone-gram units, and, in a second phase, we have used the RNNLMs to obtain the scores of each language in the identification task following a PPRLM structure. Results in terms of Cavg on the KALAKA-3 database show that the use of phone-grams provides up to 14.4% relative improvement over a baseline using only phonemes as features. In addition, our proposed strategy of reducing the number of phone-gram units using neural embeddings contributes to obtain up to 22.5% relative improvement. Finally, fusing the best system with MFCC-based acoustic i-vectors and a traditional PPRLM architecture provides up to 37.76% improvement. [ABSTRACT FROM AUTHOR]

We propose two kinds of language recognition models based on partitioned cellular automata (PCAs). They are one-dimensional PCA acceptors (PCAAs), and one-dimensional partitioned iterative array acceptors (PIAAs). We investigate how their accepting capabilities are affected by the constraint of reversibility. It is well known that (irreversible) deterministic cellular automaton acceptors and iterative array acceptors whose space is bounded by the length of the input are equivalent to deterministic linear-bounded automata (DLBAs). Here, we show reversible PCAAs and reversible PIAAs with the same space bound are also equivalent to them. These results are proved by giving concrete construction methods of a reversible PCAA and a reversible PIAA that can simulate a given DLBA. Thus, the reversibility constraint does not weaken the accepting powers of these models. [ABSTRACT FROM AUTHOR]

The manuscript provides a novel method for language identification using the texture analysis of the script. The method consists of mapping each letter from the text with certain script type. It is made according to characteristics concerning the position of the letter in the baseline area. In order to extract features, the co-occurrence matrix is computed. Then, the texture features are calculated. Extracted measures show meaningful differences due to dissimilarities in the script and language characteristics. It represents a basis in a decision-making process of the language identification. Feature classification is performed by the extension of a state-of-the-art method called genetic algorithms image clustering for document analysis. The proposed method is tested on an example of documents given in English, French, Slovenian and Serbian languages and compared to other well-known classification methods and feature representations in the state of the art. The results of experiments show the superiority of the proposed approach. [ABSTRACT FROM AUTHOR]