Machine Translation Applications Research Articles

Automatic Speech Recognition with Stuttering Speech Removal using Long Short-Term Memory (LSTM)

Stuttering or Stammering is a speech defect within which sounds, syllables, or words are rehashed or delayed, disrupting the traditional flow of speech. Stuttering can make it hard to speak with other individuals, which regularly have an effect on an individual's quality of life. Automatic Speech Recognition (ASR) system is a technology that converts audio speech signal into corresponding text. Presently ASR systems play a major role in controlling or providing inputs to the various applications. Such an ASR system and Machine Translation Application suffers a lot due to stuttering (speech dysfluency). Dysfluencies will affect the phrase consciousness accuracy of an ASR, with the aid of increasing word addition, substitution and dismissal rates. In this work we focused on detecting and removing the prolongation, silent pauses and repetition to generate proper text sequence for the given stuttered speech signal. The stuttered speech recognition consists of two stages namely classification using LSTM and testing in ASR. The major phases of classification system are Re-sampling, Segmentation, Pre-Emphasis, Epoch Extraction and Classification. The current work is carried out in UCLASS Stuttering dataset using MATLAB with 4% to 6% increase in accuracy when compare with ANN and SVM.

Quality and Machine Translation: An Evaluation of Online Machine Translation of English into Arabic Texts

This study compares the translation outputs of an English into Arabic text using the three machine translators of Google Translate, Microsoft Bing, and Ginger. To carry this evaluation of the machine translation (MT) outputs, an English text and its Arabic counterpart were selected from the UN records. The English source text was segmented into 84 semantic chunks. Depending on the Arabic counterpart model text, each chunk was rated as “correct or incorrect” at the two levels of the translation attributes: fidelity and intelligibility. To perform the quantitative description of the evaluation process, the numbers of fidelity and intelligibility errors and their percentages were calculated. Results of this evaluation process revealed that none of the three translated versions of the source text was perfectly translated. Although the translation of Microsoft Bing was rated the best, Google’s translation was found the least accurate due to the high percentage of fidelity and intelligibility errors detected in its translation output. However, the quality of Ginger’s translation was found slightly less accurate than that of Microsoft Bing, but remarkably better than Google’s translation. The findings of this study imply that these MT applications can be implemented to perform English into Arabic translation to get the broad gist of a source text, but a deep and thorough post-editing process looks essential for a full and accurate understanding of an English into Arabic MT output. The study recommends that more studies are encouraged to continue to assess the quality of MT that will further highlight its weaknesses and the strategies that should be adopted to overcome them.

More than tweets

Abstract The application of machine translation (MT) in crisis settings is of increasing interest to humanitarian practitioners. We collaborated with industry and non-profit partners: (1) to develop and test the utility of an MT system trained specifically on crisis-related content in an under-resourced language combination (French-to-Swahili); and (2) to evaluate the extent to which speakers of both French and Swahili without post-editing experience could be mobilized to post-edit the output of this system effectively. Our small study carried out in Kenya found that our system performed well, provided useful output, and was positively evaluated by inexperienced post-editors. We use the study to discuss the feasibility of MT use in crisis settings for low-resource language combinations and make recommendations on data selection and domain consideration for future crisis-related MT development.

Fast Deep Neural Network Training on Distributed Systems and Cloud TPUs

Since its creation, the ImageNet-1k benchmark set has played a significant role as a benchmark for ascertaining the accuracy of different deep neural net (DNN) models on the image classification problem. Moreover, in recent years it has also served as the principal benchmark for assessing different approaches to DNN training. Finishing a 90-epoch ImageNet-1k training with ResNet-50 on a NVIDIA M40 GPU takes 14 days. This training requires $10^{18}$1018 single precision operations in total. On the other hand, the world's current fastest supercomputer can finish $3 \times 10^{17}$3×1017 single precision operations per second (according to the Nov 2018 Top 500 results). If we can make full use of the computing capability of the fastest supercomputer, we should be able to finish the training in several seconds. Over the last two years, researchers have focused on closing this significant performance gap through scaling DNN training to larger numbers of processors. Most successful approaches to scaling ImageNet training have used the synchronous mini-batch stochastic gradient descent (SGD). However, to scale synchronous SGD one must also increase the batch size used in each iteration. Thus, for many researchers, the focus on scaling DNN training has translated into a focus on developing training algorithms that enable increasing the batch size in data-parallel synchronous SGD without losing accuracy over a fixed number of epochs. In this paper, we investigate supercomputers’ capability of speeding up DNN training. Our approach is to use a large batch size, powered by the Layer-wise Adaptive Rate Scaling (LARS) algorithm, for efficient usage of massive computing resources. Our approach is generic, as we empirically evaluate the effectiveness on five neural networks: AlexNet, AlexNet-BN, GNMT, ResNet-50, and ResNet-50-v2 trained with large datasets while preserving the state-of-the-art test accuracy. Compared to the baseline of a previous study from Goyal et al. [1] , our approach shows higher test accuracy on batch sizes that are larger than 16K. When we use the same baseline, our results are better than Goyal et al. for all the batch sizes (Fig. 20 ). Using 2,048 Intel Xeon Platinum 8160 processors, we reduce the 100-epoch AlexNet training time from hours to 11 minutes. With 2,048 Intel Xeon Phi 7250 Processors, we reduce the 90-epoch ResNet-50 training time from hours to 20 minutes. Our implementation is open source and has been released in the Intel distribution of Caffe, Facebook's PyTorch, and Google's TensorFlow. The difference between this paper and the conference-version of our work [2] includes: (1) we implement our approach on Google's cloud Tensor Processing Unit (TPU) platform, which verifies our previous success on CPUs and GPUs. (2) we scale the batch size of ResNet-50-v2 to 32K and achieve 76.3 percent accuracy, which is better than the 75.3 percent accuracy achieved in our conference paper. (3) we apply our approach to Google's Neural Machine Translation (GNMT) application, which helps us to achieves 4× speedup on the cloud TPUs.

Article retracted.

This article has been retracted by the authors.

Issues In Word Alignment From Hindi-English Languages

The paper discusses the various methods and issues related to word alignment. This paper focus on the main problem arises in word alignment because Hindi language is based on subject object verb “SVO” and for English language is subject verb objects “SOV”. Hindi is morphology rich language, therefore correct alignment of word order from Hindi to English language is quite difficult. The paper presents survey on for foreign and Indian language of word alignment in the application of machine translation

RNN-LSTM-GRU based language transformation

In past, rule-based and statistical machine translation techniques were employed to solve Urdu transliteration techniques. As mentioned in the literature, Urdu is considered as low-resource language. An impressive effort has been made for Arabic, French and Chinese language transliteration as compared to the Urdu language. Machine translation of Urdu language is a challenging problem. A very minute research work has been conducted toward Urdu transliteration. Factors behind the ignorance of Urdu language in research may be for its morphological complexity, diversity and most importantly due to the lack of reasonable bilingual parallel dataset. Getting a corpus for a language transliteration is the main resource to work on. This paper demonstrates the application of neural machine translation (NMT) for Urdu language transliteration, with the emphasis on contextual coverage of a language, which helps to improve transliteration accuracy. Build a robust NMT model which delivers efficient performance when trained over bilingual parallel corpora. Neural machine translation is an emerging technique depicting impressive performance, better than traditional MT methods in multiple aspects. In this research, we build the NMT model for the Urdu language to improve transliteration quality. An attention-based encoder–decoder system is proposed, and our experiment proves the efficiency of the proposed approach. To the best of our knowledge, this is the first effort for Urdu language bidirectional transliteration toward neural machine translation.

The undergraduate learner translator corpus: a new resource for translation studies and computational linguistics

Around the world, a growing interest has been seen in learner translator corpora, which are invaluable resources for teaching and research. This paper introduces a new resource to support researchers from different interdisciplinary areas such as computational linguistics, descriptive translation studies, computer-aided translation technology, Arabic machine translation applications, cognitive science, and translation pedagogy. Motivated by the lack of learner translator resources that provide data about learners of translation from and into Arabic, the undergraduate learner translator corpus (ULTC) is an ongoing, error-tagged sentence-aligned parallel corpus of English, Arabic, and French, with Arabic as its main language. The present corpus, consisting of parallel texts of female learners of translation from English or French into Arabic, is the first of its kind in terms of the languages represented, tasks covered, and number of students involved. It is also unique in terms of combining many complementary corpora of cross-lingual data, each of which has its own web-based query interface and corpus analysis tools. This paper describes the ULTC compilation process, preliminary findings, and planned future expansion and research.

Minimalistic Approach to Coreference Resolution in Lithuanian Medical Records.

Coreference resolution is a challenging part of natural language processing (NLP) with applications in machine translation, semantic search and other information retrieval, and decision support systems. Coreference resolution requires linguistic preprocessing and rich language resources for automatically identifying and resolving such expressions. Many rarer and under-resourced languages (such as Lithuanian) lack the required language resources and tools. We present a method for coreference resolution in Lithuanian language and its application for processing e-health records from a hospital reception. Our novelty is the ability to process coreferences with minimal linguistic resources, which is important in linguistic applications for rare and endangered languages. The experimental results show that coreference resolution is applicable to the development of NLP-powered online healthcare services in Lithuania.

FUNCTIONAL AND PRAGMATIC ADEQUACY OF JOURNALISTIC STYLE TEXTS TRANSLATION APPLYING MACHINE TRANSLATION SYSTEMS

FUNCTIONAL AND PRAGMATIC ADEQUACY OF JOURNALISTIC STYLE TEXTS TRANSLATION APPLYING MACHINE TRANSLATION SYSTEMS

‘Webbish writing’ for global communication in the Web 3.0

Abstract Advances in information technology and the needs of globalization made English the lingua franca (EFL) of the Web 1.0. Later, multi-modes of computer-mediated communication in the Web 2.0 fostered hybrid English as a new EFL by combing English with other languages. Moving into the Web 3.0 for information mining and knowledge acquisition without a language barrier, the author proposes ‘Webbish’, a controlled form of English, to write web texts for the purpose of gisting through the production of multilingual machine translation (MT) outputs. To demonstrate the effectiveness of ‘Webbish’, an empirical survey was conducted to measure the comprehensibility of the MT of ‘Webbish’ texts based on a 1–100 grading scale. An average score of 89.9 was earned from twenty-four international participants, showing a high degree of MT comprehensibility. Meanwhile, set within the framework of postmodernism, ‘Webbish’ writing takes on postmodernist significance of non-fixed norms, dynamic meanings and an equal right of comprehensible Web information access for global non-English audiences. Its linguistic features are suited to MT application and its fluid meanings are produced due to the MT system’s rendition and different target audiences’ different interpretations. ‘Webbish’ writing suggests acceptance, inclusiveness and equality. It is also an easy way to influence and improve global communication.

ПРОБЛЕМАТИКА ПУБЛИКАЦИИ ПЕРЕВОДНЫХ СТАТЕЙ ПО ХИМИИ В АНГЛОЯЗЫЧНЫХ НАУЧНЫХ ЖУРНАЛАХ

The article considers the relevant and growing tendency of the publication of scientific research results in the English-language journals in recent years. The priority of fundamental areas of research is shown by the particular example of chemistry and adjacent natural-science branches. This research scopes the acute problems in translating chemistry articles into English. They are shown from the scientist’s position, viz. language mastering, skills to solve communicative problems with the English native speakers, typical translation mistakes. Also, the paper considers the question of rising anglicizing of scientific vocabulary, pluses and minuses of translators’ work not related to the field of study, and the application of machine translation.

Input Method for Human Translators

Computer-aided translation (CAT) systems are the most popular tool for helping human translators efficiently perform language translation. To further improve the translation efficiency, there is an increasing interest in applying machine translation (MT) technology to upgrade CAT. To thoroughly integrate MT into CAT systems, in this article, we propose a novel approach: a new input method that makes full use of the knowledge adopted by MT systems, such as translation rules, decoding hypotheses, and n-best translation lists. The proposed input method contains two parts: a phrase generation model, allowing human translators to type target sentences quickly, and an n-gram prediction model, helping users choose perfect MT fragments smoothly. In addition, to tune the underlying MT system to generate the input method preferable results, we design a new evaluation metric for the MT system. The proposed input method integrates MT effectively and imperceptibly, and it is particularly suitable for many target languages with complex characters, such as Chinese and Japanese. The extensive experiments demonstrate that our method saves more than 23% in time and over 42% in keystrokes, and it also improves the translation quality by more than 5 absolute BLEU scores compared with the strong baseline, i.e., post-editing using Google Pinyin.

(Machine translation from computerized Linguistic perspective: comparative study between (Google Translate and Microsoft Translator

Machine translation is considered one the important subjects included in Linguistic Studies in general and computerized Linguistic in particular. Several Machine translation applications have emerged in the past years and the current research focuses on identifying the term of Machine translation using several texts and comparing the translation results of two famous Machine translation applications (Microsoft and Google translator). The results showed that Machine translation is one the modern terms emerged in Linguistic field, and many improvements has been conducted on Machine translation over the years in order to improve its results, hence, the findings indicated that human translation is more accurate and there are some defects in Machine translation and results of the current study approved theses defects by the translation model used in the study.

Calculation of Semantic Distances Between Words: From Synonymy to Antonymy

A new approach to numerically measure the semantic distances between lexical units (words and collocations) based on the geometric analogies and analytical calculation, is put forward. Having considered the cases of equal and different weights of semes, we obtained exact algebraic formulas describing different levels of the meanings proximity, ranging from absolute synonymy to full antonymy. It was emphasized that absolute synonymy arises when the compared units contain equal numbers of semes that fully coincide and have equal weights in the corresponding pairs. Calculation of the semes weights helps to locate the unit more precisely on the semantic sphere. It was shown that the level of synonymy and antonymy decreases if different semes are accentuated, while the semantic distance between the units without identical semes cannot be influenced by seme boosting. It was observed that depending on the context, a word may wander over this sphere, thus modifying its lexical semantic relations with other units. As the proposed approach contributes to formalization of the units comparison procedure, it is advisable for incorporation into relevant automatic tools, particularly into WordNet and FrameNet. The obtained results may be useful for various linguistic and associated studies including automatic text analysis and processing, computer lexicography, information search and retrieval, machine translation and other NLP applications that are related to the artificial intelligence problem.

Long Text Generation via Adversarial Training with Leaked Information

Automatically generating coherent and semantically meaningful text has many applications in machine translation, dialogue systems, image captioning, etc. Recently, by combining with policy gradient, Generative Adversarial Nets(GAN) that use a discriminative model to guide the training of the generative model as a reinforcement learning policy has shown promising results in text generation. However, the scalar guiding signal is only available after the entire text has been generated and lacks intermediate information about text structure during the generative process. As such, it limits its success when the length of the generated text samples is long (more than 20 words). In this paper, we propose a new framework, called LeakGAN, to address the problem for long text generation. We allow the discriminative net to leak its own high-level extracted features to the generative net to further help the guidance. The generator incorporates such informative signals into all generation steps through an additional MANAGER module, which takes the extracted features of current generated words and outputs a latent vector to guide the WORKER module for next-word generation.Our extensive experiments on synthetic data and various real-world tasks with Turing test demonstrate that LeakGAN is highly effective in long text generation and also improves the performance in short text generation scenarios. More importantly, without any supervision, LeakGAN would be able to implicitly learn sentence structures only through the interaction between MANAGER and WORKER.

Tecnologias e formação de tradutores

Os sistemas de memórias de tradução e as novas aplicações de programas de tradução automática são parte do cotidiano de tradutores que trabalham com traduções especializadas e que prestam serviços para a indústria da localização. A integração dessas tecnologias está promovendo mudanças em muitos cursos universitários de formação de tradutores que, gradativamente, estão revendo seus currículos para incluir o treinamento em ferramentas. Este trabalho relata resultados de dois estudos conduzidos em dois cursos de graduação em tradução em duas universidades públicas brasileiras. Os resultados apontam para a importância de se incluírem, nas aulas de prática de tradução, discussões que estimulem o desenvolvimento de um olhar crítico em relação às ferramentas que os tradutores utilizam, para que aprendam a valorizar seu papel como negociador de sentidos em todo trabalho de tradução.

Metadata records machine translation combining multi‐engine outputs with limited parallel data

One way to facilitate Multilingual Information Access (MLIA) for digital libraries is to generate multilingual metadata records by applying Machine Translation (MT) techniques. Current online MT services are available and affordable, but are not always effective for creating multilingual metadata records. In this study, we implemented 3 different MT strategies and evaluated their performance when translating English metadata records to Chinese and Spanish. These strategies included combining MT results from 3 online MT systems (Google, Bing, and Yahoo!) with and without additional linguistic resources, such as manually‐generated parallel corpora, and metadata records in the two target languages obtained from international partners. The open‐source statistical MT platform Moses was applied to design and implement the three translation strategies. Human evaluation of the MT results using adequacy and fluency demonstrated that two of the strategies produced higher quality translations than individual online MT systems for both languages. Especially, adding small, manually‐generated parallel corpora of metadata records significantly improved translation performance. Our study suggested an effective and efficient MT approach for providing multilingual services for digital collections.

Linguistic evaluation of translation errors in Chinese–English machine translations of patent titles

Abstract The title of an invention allows the reader to understand the significance of a patent claim, and the wording of the title recurs throughout the subsequent patent documentation. If the translation of the title is erroneous, the quality of the translation in other parts of the patent documentation also suffers. This research involved using linguistic evaluation to identify common translation errors in Chinese–English machine translations of patent titles and examine the quality of machine-translated patent titles. Special focus was placed on orthographic, morphological, lexical, semantic, and syntactic errors found in patent titles. We sought to answer the following questions: (1) What are the trends in the application of machine translation in the Taiwan Intellectual Property Office (TIPO)? (2) How is the quality of machine translation controlled at TIPO? (3) What are common errors in machine-translated patent titles? Through analysis of our findings, it is possible to estimate the level of effort required from a posteditor following translation, and to suggest methods of improving machine translations of patent titles. This study also generates information applicable to the training of patent translators and posteditors.

Analysis of the Current Development of Machine Translation and Interpretation in Korea: Focusing on Korean-Chinese Language Pairs

With the application of artificial intelligence technologies, machine translation and interpretation has recently made rapid progress. In Korea, Naver Corp., Hancom Interfree Inc., and Systran International Co. Ltd. took lead in providing desktop- and mobile-based translation and interpreting services. All three enterprises apply neural machine translation (NMT), currently the latest technology in machine translation. However, they have different characteristics: Naver, the largest Web portal site in Korea, quickly responds to colloquial expressions or newly coined words by using various types of content as data for translation engine. Hancom Interfree developed Genietalk; its hybrid translation engine combines the strengths of NMT and rule-based machine translation (RBMT), and thus enhances the accuracy of translation. Capitalizing on the advantages of a translation software company, Systran features expertise of translation by industry and area. These ever-evolving machine translation and interpretation are expected to satisfy daily demands for interpretation in shopping and tourism, demands for community interpretation from foreigners living in Korea including marriage immigrants, and temporary demands for translation and interpretation from international events including Olympics. However, in professional areas, it is hard to replace human translation and interpretation with machines in a short term. Rather, the efficiency of translation and interpreting jobs is expected to improve through cooperation between human beings and machines.