Evaluation of 2-way Iraqi Arabic–English speech translation systems using automated metrics

Condon, Sherri L.; Arehart, Mark; Parvaz, Dan; Sanders, Gregory A.; Doran, Christine; Aberdeen, John S.

doi:10.1007/s10590-011-9105-x

Cited by 8 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reference [8] shows that conducted in Extending BLEU Evaluation Method with Linguistics Weight research to improve the effectiveness of BLEU method, and they succeeded through the use of multiple Ngram weights. Reference [9] evaluates 2-way Iraqi Arabic-English speech translation systems using automated metrics. They found that automatic translation of Iraqi Arabic correlates with human judgements.…”

Section: Review Of Relevant Literaturementioning

confidence: 99%

Assessing the Translation of Google and Microsoft Bing in Translating Political Texts from Arabic into English

Almahasees¹

2017

IJLLL

View full text Add to dashboard Cite

Abstract-Online machine translation (OMT) systems are widely used throughout the world freely or at low cost. Most of these systems use statistical machine translation (SMT) that is based on a corpus full with translation examples to learn from them how to translate correctly. Online automatic machine translation systems differ widely in their effectiveness and accuracy. Therefore, the wide spread of such translation platforms make it necessary to evaluate the output in order to shed light on the capacity and usability of each system. The present study have selected the most prominent translation systems, Google and Microsoft to test which system is better and more reliable in rendering English<>Arabic translation. To conduct the study, the researcher has chosen automatic evaluation of the two system outputs by using the most popular automatic evacuation metric BLEU. The study's corpus consists of 25 Arabic sentences extracted from Petra News Agency of Jordan with its human reference translation from the English version of Petra. The result of the research showed that Google translate achieves better results than Microsoft Bing in comparison to human referenced translation. However, Machine Translation (MT) is still far from reaching fully automatic translation of a quality obtained by human translators.

show abstract

Section: Review Of Relevant Literaturementioning

confidence: 99%

Assessing the Translation of Google and Microsoft Bing in Translating Political Texts from Arabic into English

Almahasees¹

2017

IJLLL

View full text Add to dashboard Cite

show abstract

“…Motoyuki and Kazuhiro discussed traditional machine translation methods based on dictionaries and conversion rules and example machine translation methods based on parallel corpora [11]. Condon et al believe that any language problems faced by translation can be attributed to word meaning problems at the lexical level and structure problems at the grammatical level [12]. Wen et al put forward a diagonal alignment model and an alignment model based on hidden Markov model.…”

Section: Research On Machine Translation Technologymentioning

confidence: 99%

English Long Sentence Segmentation and Translation Optimization of Professional Literature Based on Hierarchical Network of Concepts

2022

Mobile Information Systems

View full text Add to dashboard Cite

The structure of English long sentences is generally complicated, with complicated logical levels, many parallel elements, many modifiers and conjunctions, and long post attributives. Moreover, pronouns in English long sentences often need to be judged according to the context. Complex English long sentences are very common in English and even run through the whole English article. The analysis results of these complex long sentences will seriously affect the quality and readability of machine translation. In this paper, HNC (hierarchical network of concepts) method is improved to realize the segmentation of long sentences, so as to simplify sentence patterns. According to the characteristics of professional literature, this paper puts forward a translation optimization method of professional literature combining HNC with statistics, which greatly improves the recognition efficiency of unknown words by extracting professional terms. The results show that the word segmentation method based on HNC and statistics proposed in this paper has achieved a good word segmentation effect in the open test environment, with an accuracy rate of 93.38% and a recall rate of 94.51%. The conclusion shows that our method can make full use of the knowledge of the source tree database, thereby improving the accuracy of the syntactic model on the target tree database.

show abstract

“…Arabs use many dialects that vary from a place to another, and Iraqi Arabic is one of these dialects that used in Iraq as the name indicates, and it is close to dialects used in the gulf region. Condon et al study [24] presents an automatic method to evaluate the quality of Iraqi Arabic-English speech translation dialogues. They show that normalization has a positive effect on making the candidate translations closer to human judgments.…”

Section: Related Workmentioning

confidence: 99%

Comparative Study Between METEOR and BLEU Methods of MT: Arabic into English Translation as a Case Study

Hadla¹,

Hailat²,

Al-Kabi³

2015

ijacsa

View full text Add to dashboard Cite

Abstract-The Internet provides its users with a variety of services, and these services include free online machine translators, which translate free of charge between many of the world's languages such as Arabic, English, Chinese, German, Spanish, French, Russian, etc. Machine translators facilitate the transfer of information between different languages, thus eliminating the language barrier, since the amount of information and knowledge available varies from one language to another, Arabic content on the internet, for example, accounts 1% of the total internet content, while Arabs constitute 5% of the population of the earth, which means that the intellectual productivity of the Arabs is low because within internet use Internet's Arabic content represents 20% of their natural proportion, which in turn encouraged some Arab parties to improve Arabic content within the internet. So, many of those interested specialists rely on machine translators to bridge the knowledge gap between the information available in the Arabic language and those in other living languages such as English.This empirical study aims to identify the best Arabic to English Machine translation system, in order to help the developers of these systems to enhance the effectiveness of these systems. Furthermore, such studies help the users to choose the best. This study involves the construction of a system for Automatic Machine Translation Evaluation System of the Arabic language into language. This study includes assessing the accuracy of the translation by the two known machine translators, Google Translate, and the second, which bears the name of Babylon machine translation from Arabic into English. BLEU and METEOR methods are used the MT quality, and to identify the closer method to human judgments. The authors conclude that BLEU is closer to human judgments METEOR method.

show abstract

Evaluation of 2-way Iraqi Arabic–English speech translation systems using automated metrics

Cited by 8 publications

References 8 publications

Assessing the Translation of Google and Microsoft Bing in Translating Political Texts from Arabic into English

Assessing the Translation of Google and Microsoft Bing in Translating Political Texts from Arabic into English

English Long Sentence Segmentation and Translation Optimization of Professional Literature Based on Hierarchical Network of Concepts

Comparative Study Between METEOR and BLEU Methods of MT: Arabic into English Translation as a Case Study

Contact Info

Product

Resources

About