Translation System Research Articles

Psychological Paradigm of the Implementation of Awakening-Motivational, Analytical-Synthetic and Executive Phases into the Process of Translation Activity

The purpose of our research is to reveal the ways of the implementation of awakening-motivational, analytical-synthetic and executive phases into the process of translation activity. Methods of the research. The following theoretical methods of the research were used to solve the tasks formulated in the article: the categorical method, structural and functional methods, the methods of the analysis, systematization, modeling and generalization. The ascertaining research was used as an empirical method. The results of the research. Analyzing speech activity, we’d like to emphasize that each individual act of the translation activity begins with a motive and planning. Also, each act ends with a result, the achievement of a certain goal, and the basis of the structure of the translation activity. In such a way, translation activity is a dynamic system of specific actions and operations, which have the aim to be the result of this achievement. Such a dynamic system of Translation is the basis of the operational and prognostic mechanisms of the translator’s activity; it largely determines the speed and automatism of translation activities in the whole. The psychological content of translation activity includes different conditions of this activity, in which this activity takes a place. In such a way translation activity is determined by such elements, as the subject, means, tools, product and result. Conclusions. The subject of the translational activity determines its nature. It is precisely in it that the need of providing translational activity is realized. For example, in oral translational activity the subject is thought as a form of reflection of the relationships between objects and phenomena of a natural reality. It is in the process of expressing thoughts that the purpose of speech lies, in particular, in the reproduction of thoughts by the translator. The purpose of listening is realized, in turn. At the same time, a foreign language is a means of social communication, a means of expression and understanding, a tool of translational activity. It is a thought as an element of the psychological content of oral translational activity that determines the features and the conditions of its implementation.

Neural machine translation systems for English to Khasi: A case study of an Austroasiatic language

Neural machine translation has predominantly outperformed previous machine translation models primarily for resourceful languages. However, very little work has been reported for resource-constrained languages such as Khasi. The Khasi language belongs to the Mon-Khmer branch of the Austroasiatic language family and is spoken primarily in the state of Meghalaya in India. Although performing neural machine translation for the under-resourced Khasi language is difficult, we build a substantial parallel corpus of English–Khasi. We apply three segmentation methods in the datasets for our experiments: untokenized, tokenized and a subword BPE (Byte Pair Encoding). Experiments are carried out on this dataset with different aspects of neural machine translation systems using cutting-edge architectures such as LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit) and a transformer based model for the English–Khasi language pair. We also carry out experiments by adapting the transfer learning approach using English-Vietnamese as the parent language pair and English-Khasi as the child language pair. This work reports a quantitative and qualitative analysis of several models based on architectural and data segmentation methodologies. The experimental findings show that the model adapted using the transfer learning approach achieved a reasonable improvement in BLEU scores with the highest being 58.1 BLEU on similar domains and 17.7 BLEU for the general domain outscoring the other models for the same language pair. Qualitative analysis is carried out focusing on the morphological inflections of gender identification in the translated output.

North Korean Neural Machine Translation through South Korean Resources

South and North Korea both use the Korean language. However, Korean natural language processing (NLP) research has mostly focused on South Korean language. Therefore, existing NLP systems in the Korean language, such as neural machine translation (NMT) systems, cannot properly process North Korean inputs. Training a model using North Korean data is the most straightforward approach to solving this problem, but the data to train NMT models are insufficient. To solve this problem, we constructed a parallel corpus to develop a North Korean NMT model using a comparable corpus. We manually aligned parallel sentences to create evaluation data and automatically aligned the remaining sentences to create training data. We trained a North Korean NMT model using our North Korean parallel data and improved North Korean translation quality using South Korean resources such as parallel data and a pre-trained model. In addition, we propose Korean-specific pre-processing methods, character tokenization, and phoneme decomposition to use the South Korean resources more efficiently. We demonstrate that the phoneme decomposition consistently improves the North Korean translation accuracy compared to other pre-processing methods.

On the challenges of legal translation

This short commentary focuses on language and law and legal translation. It stresses the importance of language used in legal texts in the legal context and the complexities when more than one language is involved in interlingual and cross-cultural communication in law. There are age old challenges in legal translation as well as new ones in the digital age with the increasing use of machine translation systems. It is imperative today that human legal translators are familiar with machine translation tools and aware of how computer aided translation technologies process information and their strengths and weaknesses. To be able to effectively use machine translation systems should become a compulsory part of the digital literacy and skill sets of legal translators in the twenty-first century.

Retracted: English Language Intelligent Translation System Based on 3D Visualization Technology

Retracted: English Language Intelligent Translation System Based on 3D Visualization Technology

An empirical analysis on statistical and neural machine translation system for English to Mizo language

An empirical analysis on statistical and neural machine translation system for English to Mizo language

Design of a Smart Teaching English Translation System Based on Big Data Machine Learning

In the context of artificial intelligence, the use of machine translation in English reading classroom teaching is a more common learning method. In traditional teaching methods, machine translation is more convenient and faster than human translation, but it often deviates from the original text in terms of grammar and sentence pattern. Based on the perspective of English reading class, this paper compares traditional and machine translation, and discusses the future development trend and influence mechanism of the current situation of using machine translation in English reading class under the effect of artificial intelligence.

Towards better Chinese-centric neural machine translation for low-resource languages

The last decade has witnessed enormous improvements in science and technology, stimulating the growing demand for economic and cultural exchanges in various countries. Building a neural machine translation (NMT) system has become an urgent trend, especially in the low-resource setting. However, recent work tends to study NMT systems for low-resource languages centered on English, while a few studies focus on low-resource NMT systems centered on other languages such as Chinese. To achieve this, the low-resource multilingual translation challenge of the 2021 iFLYTEK AI Developer Competition provides the Chinese-centric multilingual low-resource NMT tasks, where participants are required to build NMT systems based on the provided low-resource samples. In this paper, we present the competition-winning system that leverages monolingual word embeddings data enhancement, bilingual curriculum learning, and contrastive re-ranking. In addition, a new Incomplete-Trust (In-trust) loss function is proposed to replace the traditional cross-entropy loss when training. The extensive experimental results demonstrate that the implementation of these ideas leads to better performance than other state-of-the-art methods. All the experimental codes are released at: https://github.com/WENGSYX/Low-resource-text-translation.

Sentiment induced phrase-based machine translation: Robustness analysis of PBSMT with senti-module

Every type of machine translation system (i.e. neural, statistical, rule-based machine translation system) is equal important to build a sophistical hybrid machine translation system. Keeping this fact in my mind, I concentrate to improve statistical machine translation system with more natural way. In this paper, I try to preserve sentiment after translation to improve the overall accuracy of the machine translation system. So, I introduced senti-model here. A senti-model (sentiment model), translation model, language model, and distortion model are incorporated on the top of the beam search algorithm for decoding. At first, sentiment information is learned and modeled with translation probability by using this algorithm. Thereafter, I decode the source sentences-based on the contextual information. Overall procedure of translation modeling with a sentiment, parameter estimation for it, and senti-translation decoding (decoding with the sentiment model) are presented with empirical evidence. Experiments on a benchmark English–Hindi dataset shows that the proposed model is capable to improve the accuracy (in terms of 4.66 BLEU points, 4.09 LeBleu points, 4.67 NIST points, 5.71 RIBES points) significantly and preserves sentiment 7.79% more than the state-of-the-art technique.

Algorithmic Translation Correction Mechanisms: An End-to-end Algorithmic Implementation of English-Chinese Machine Translation

INTRODUCTION: Machine translation is a modern natural language processing research field with important scientific and practical significance. In practice, the variation of languages, the limitation of semantic knowledge, and the lack of parallel language resources limit the development of machine translation. OBJECTIVES: This paper aims to avoid duplicating neural networks during the learning process and improve the ability to generalize complex neural network machine translation models with limited resources. METHODS: Textual material in the source language was studied, and a suitable textual material representation model was used to express complex, high-level, and abstract semantic information. Then, a more efficient neural network machine translation integration model was developed based on the control of written data and algorithms. RESULTS: Data mining must be applied to complex neural network machine translation systems based on transfer learning to standardize finite neural network models. CONCLUSION: Neural network-based embedded machine translation systems based on migration training require a small number of labelled samples to improve the system's permeability. However, this adaptive migration learning region approach can easily lead to over-learning problems in neural network machine translation models, thus avoiding excessive correspondences during the learning process and improving the generalization ability of the translation model with limited neural network resources.

Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case

Recently, there has been a growth in the number of studies addressing the automatic processing of low-resource languages. The lack of speech and text data significantly hinders the development of speech technologies for such languages. This paper introduces an automatic speech recognition system for Livvi-Karelian. Acoustic models based on artificial neural networks with time delays and hidden Markov models were trained using a limited speech dataset of 3.5 h. To augment the data, pitch and speech rate perturbation, SpecAugment, and their combinations were employed. Language models based on 3-grams and neural networks were trained using written texts and transcripts. The achieved word error rate metric of 22.80% is comparable to other low-resource languages. To the best of our knowledge, this is the first speech recognition system for Livvi-Karelian. The results obtained can be of a certain significance for development of automatic speech recognition systems not only for Livvi-Karelian, but also for other low-resource languages, including the fields of speech recognition and machine translation systems. Future work includes experiments with Karelian data using techniques such as transfer learning and DNN language models.

Exploration of low-resource language-oriented machine translation system of genetic algorithm-optimized hyper-task network under cloud platform technology

Exploration of low-resource language-oriented machine translation system of genetic algorithm-optimized hyper-task network under cloud platform technology

A Review of Machine Translation Quality Assessment Methods

Machine translation quality assessment plays a crucial role in enhancing the performance of machine translation systems. This review aims to survey and outline the current major methods used for assessing machine translation quality, which can be broadly categorized into manual and automatic assessment methods. Upon analyzing the existing literature, it becomes evident that while manual evaluation methods yield high-quality results, they are time-consuming and labor-intensive. On the other hand, automatic evaluation methods are cost-effective and rapid, but their evaluation results do not match the expertise of human evaluators. The objective of this paper is to furnish researchers with a comprehensive overview of machine translation quality assessment methods, enabling them to select appropriate approaches based on their specific experimental requirements. Additionally, we aspire to offer valuable insights and novel perspectives for further advancements in the field of machine translation quality assessment methods.

Automatic Resource Augmentation for Machine Translation in Low Resource Language: EnIndic Corpus

Parallel corpus is the primary ingredient of machine translation. It is required to train the statistical machine translation (SMT) and neural machine translation (NMT) systems. There is a lack of good quality parallel corpus for Hindi to English. Comparable corpora for a given language pair are comparatively easy to find, but this cannot be used directly in SMT or NMT systems. As a result, we generate a parallel corpus from the comparable corpus. For this purpose, the sentences (which are translations of each other) are mined from the comparable corpus to prepare the parallel corpus. The proposed algorithm uses the length of the sentence and word translation model to align sentence pairs that are translations of each other. Then, the sentence pairs that are poor translations of each other (measured by a similarity score based on IBM model 1 translation probability) are filtered out. We apply this algorithm to comparable corpora, which are crawled from speeches of the President and Vice-President of India, and mined parallel corpora out of them. The prepared parallel corpus contains good quality aligned sentences (with 96.338% f-score). Subsequently, incorrect sentence pairs are filtered out manually to make the corpus in qualitative practical use. Finally, we gather various sentences from different sources to prepare the EnIndic corpus, which comprises 1,656,207 English-Hindi sentence pairs (miscellaneous domain). We have deployed this prepared largest English-Hindi parallel corpus at https://github.com/debajyoty/EnIndic.git and the source code at https://github.com/debajyoty/EnIndicSourceCode.git.

Drosophila melanogaster as a Translational Model System to Explore the Impact of Phytochemicals on Human Health.

Fruits, vegetables, and spices are natural sources of bioactive phytochemicals, such as polyphenols, carotenoids, flavonoids, curcuminoids, terpenoids, and capsaicinoids, possessing multiple health benefits and relatively low toxicity. These compounds found in the diet play a central role in organism development and fitness. Given the complexity of the whole-body response to dietary changes, invertebrate model organisms can be valuable tools to examine the interplay between genes, signaling pathways, and metabolism. Drosophila melanogaster, an invertebrate model with its extensively studied genome, has more than 70% gene homology to humans and has been used as a model system in biological studies for a long time. The notable advantages of Drosophila as a model system, such as their low maintenance cost, high reproductive rate, short generation time and lifespan, and the high similarity of metabolic pathways between Drosophila and mammals, have encouraged the use of Drosophila in the context of screening and evaluating the impact of phytochemicals present in the diet. Here, we review the benefits of Drosophila as a model system for use in the study of phytochemical ingestion and describe the previously reported effects of phytochemical consumption in Drosophila.

Machine translation of standardised medical terminology using natural language processing: A scoping review

Standardised medical terminologies are used to ensure accurate and consistent communication of information and to facilitate data exchange. Currently, many terminologies are only available in English, which hinders international research and automated processing of medical data. Natural language processing (NLP) and Machine Translation (MT) methods can be used to automatically translate these terms. This scoping review examines the research on automated translation of standardised medical terminology. A search was performed in PubMed and Web of Science and results were screened for eligibility by title and abstract as well as full text screening. In addition to bibliographic data, the following data items were considered: 'terminology considered', 'terms considered', 'source language', 'target language', 'translation type', 'NLP technique', 'NLP system', 'machine translation system', 'data source' and 'translation quality'. The results showed that the most frequently translated terminology is SNOMED CT (39.1%), followed by MeSH (13%), ICD (13%) and UMLS (8.7%). The most common source language is English (55.9%), and the most common target language is German (41.2%). Translation methods are often based on Statistical Machine Translation (SMT) (41.7%) and, more recently, Neural Machine Translation (NMT) (30.6%), but can also be combined with various MT methods. Commercial translators such as Google Translate (36.4%) and automatic validation methods such as BLEU (22.2%) are frequently used tools for translation and subsequent validation.

Improving Access to Medical Information for Multilingual Patients using Pipelined Ensemble Average based Machine Translation

Machine translation has shown potential in improving access to medical information and healthcare services for multilingual patients. This research aims to enhance machine translation accuracy in the medical field, specifically for translating from Hindi to English. The study introduces a new approach that dynamically allocates decoding parameters using regression models, overcoming the limitations of fixed parameters in the decoder. A comprehensive dataset is created to address limited data availability, enabling regression models to predict optimal pruning parameters. The main motivation for the study is the introduction of a regression method for optimizing pruning parameters, which is a novel approach in this context. The proposed approach outperforms existing methods, achieving improved translation accuracy. Standard metrics such as the BLEU score are used to evaluate translations. Ensemble average and pipeline approaches further enhance performance. The improved performance of the proposed models can be attributed to the ensemble of diverse models (Extra Trees, LightGBM, XGBoost, and Random Forest) that employ various techniques to reduce overfitting, enhance prediction accuracy, and improve translation by correcting prediction errors. The study contributes to facilitating the translation and sharing of medical literature, promoting collaboration and knowledge exchange across languages. The research demonstrates the effectiveness of the regression method for optimizing pruning parameters in machine translation, leading to improved translation accuracy in the medical field. The proposed models offer promising results, paving the way for enhanced machine translation systems and promoting collaboration and knowledge exchange in the medical domain. The source code is available at https://huggingface.co/debajyoty/statistical-regression-Based-MT/tree/main/Statistical-Regression-SMT.

The development of a wearable device for Sign Language Gesture Recognition (SLGR) by using Conceptual Design Approach (CDA)

In recent years, there has been a significant focus on researching and developing Sign Language Gesture Recognition (SLGR) for people with hearing and speaking impairments. This is especially important in Indonesia, where there are approximately two million people with these disabilities. However, current research and developments on SLGR devices are specifically designed to understand only particular sign language systems, such as Korean Standard Sign Language (KSDSL) and American Sign Language (ASL), each of which has its unique gestures and models. As a result, no device has been developed to recognize the gestures of the Indonesian sign language system, known as Sistem Isyarat Bahasa Indonesia (SIBI). Therefore, this study aims to develop an SLGR device that can recognize and translate SIBI gestures into output images via text and speech. The development of the SLGR device in this study is conducted by using the Conceptual Design Approach (CDA) methodology. Where in this case, previous research on SLGR devices is first observed as a benchmark for comparison. Furthermore, the benchmark is used as the basis for the function, sub-function, and specification of the proposed design. Based on these stages of benchmarking and concept development, the study concluded that the final design of the proposed SLGR device is constructed by using 5 flex and Gyroscope sensors connected wirelessly to the Raspberry microcontroller. The device is equipped with a voice system and LCR RPi as the output system for translation. Based on the combination of these sensors, the device is able to identify any particular gestures that correspond to words and phrases in SIBI and translate them into speech via the designated speaker and text display on the LCD screen. To fully understand the performance of the device, experimental tests are conducted by analyzing the input of 26 alphabets in the SIBI system. As a result, the device demonstrated an average of 92% accuracy to convert sign language into voice and text, which demonstrates the usefulness of the proposed device

Comparing the Performance of Google Translate and SYSTRAN on Arabic Lexical Ambiguity

Machine translation systems face many challenges when dealing with Arabic lexical ambiguity, which affects the quality and accuracy of their translations. This study investigates how two popular MT systems, Google Translate and SYSTRAN, handle three problematic linguistic features of the Arabic language: homonyms, heteronyms, and polysemes. A test suite was designed to include sentences that contain these features in different contexts and domains. The translations produced by the two MT systems were evaluated by four independent evaluators for intelligibility and accuracy using a four-point scale. Results showed that both MT systems struggled with the three chosen linguistic features, with average scores below 40%. Google Translate outperformed SYSTRAN in almost every sentence in the test suite. Both systems scored better in intelligibility than in accuracy. Heteronyms proved to be the most challenging for both MT systems due to the unique design of Arabic discretization, which is not yet recognized by translation systems. This study contributes to the field of machine translation by providing a comprehensive analysis of Arabic lexical ambiguity and its impact on MT quality, as well as suggesting possible ways to improve MT systems for Arabic-English translation.

Ferric Ion‐Driven Chlorin E6 Nanoparticles: Simple and Effective Multimodal Theranostics

AbstractTheranostics integrating therapeutic power and imaging exhibit great prospects in precision medicine. It is a big challenge to develop simple and selective theranostic systems for potential clinical translation. Herein the Fe (III) driven assembly of chlorin E6 (Ce6) is reported to form multifunctional nanoparticles. The resulting Fe‐Ce6 NPs possess high loading content of photosensitizer Ce6, red‐shifted absorption, good colloidal stability, and photostability, and robust photothermal performance. After cellular internalization, Fe‐Ce6 NPs can respond to intracellularly enriched glutathione to release Ce6, thus priming photodynamic therapy (PDT) effect, fluorescence, and afterglow luminescence. In vitro and in vivo results prove that Fe‐Ce6 NPs remarkably enhance the tumor regression through photothermal therapy and in situ PDT. This study provides a facile and efficient method to construct versatile nanotheranostics for tumor treatment.