Using Google Translate Effectively and Efficiently in Translating Vietnamese Texts into English Ones

Abstract

With the boom and fast development of AI, using machine translation in translation teaching and learning has been a hot issue that appeals to much attention. Google Translate, one of the most popular machine translation tools, is widely used in translation process in Vietnam university context for different reasons. At Van Lang University, where students majoring in Translation and Interpreting are discouraged to use Google Translate, they still resort to it. However, it seemed that this application did not create a high-quality product when the source language is Vietnamese. Therefore, the author tried instructing the students using Google Translate tool at the second stage after the stage of human intervention so that they can do their translation tasks effectively and efficiently. This article reports the findings from the research done with students majoring in Translation and Interpreting. It initially provides different views of difficulties in dealing with Vietnamese texts and the limitations of Google Translate in providing an English translation version. Then, it proceeds with describing the setting of the present investigation. Based on qualitative research, the information is collected and analyzed, and then, the findings are drawn. The article concludes with modest suggestions and recommendations for using Google Translate to teach translation.

1. Introduction

The Fourth Industrial Revolution with the advent of cyber physical systems has opened up a wealth of opportunity for the translation industry. The application of advances of technology and machine translation tools in unprofessional and professional translation to produce an enormous volume of translation texts in very short time has proved the fact that using machine translation is inevitable. Translation teaching and learning therefore is forced to makes changes involving instructing learners on how to use machine translation tools effectively and efficiently to produce high-quality translation versions.

Google Translate (GT), one of the most popular machine translation tools with the capability of translation of more than 100 languages in the world, is increasingly widespread. In the context of higher education in Vietnam, GT is considered a dominant translation tool in learning English for both English major students and English nonmajor students. According to the research at Hanoi University of Industry in 2021 done by Nguyen et al. [1], 100 percent of English nonmajor students surveyed used GT for learning E for Specific Purpose (ESP). The author of this study also found the similar number of English major students using GT in their learning process despite not having conducted an official survey.

English nonmajor students use GT mainly for reading and comprehending their course books written in English or learning ESP, while English major students resort to GT to deal with their courses in the specialism stage (in the university curriculum), particularly translation assignments in their translation courses. Having observed English major students’ using GT translating texts from Vietnamese into English and vice versa, the author found that employing GT in the first step and human modification in the final step in translating texts from Vietnamese into English seemed ineffective and inefficient, while the machine translation method still helped English-Vietnamese translation to some extent. The reality put the author into a question that what should be done to help the learners using GT effectively and efficiently and improving their translating techniques translating texts from Vietnamese into English instead of discouraging them from it.

The research, then, focuses on answering the two questions:

1. What difficulties may learners have when translating a text from Vietnamese to English?

2. How can GT be used effectively and efficiently to produce high-quality translation versions from Vietnamese to English?

2. Theoretical framework

2.1 Translation

Translation is a process of written communication between the writer and the possible audience through the translator’s conveying meaning of a source language into a target language considering the elements of lexicon, grammar, and context to meet the requirement of naturalness of a target text.

According to Newmark [2], translation is “rendering the meaning of a text into another language in a way that the author intended the text”. Translation is a procedure with four levels: (1) the SL text level, the level of language; (2) the referential level, the level of objects and events, real or imaginary; (3) the cohesive level; and (4) the level of naturalness.

Holding a similar viewpoint to Newmark’s, Zafitri [3] thinks that translation is the processing of replacing meanings in one language into another language through interpretations. When a text is translated, the elements of lexicon, grammar, and context in the source language should be considered and transferred into the target and the meaning of the target language text should feature the naturalness of the target language. Therefore, it is required that the equivalence between source language and target language be met in order to prevent the audience from misunderstanding the target text.

Nida [4] defined translation as a process that can be viewed from different perspectives such as stylistics, authors intent, diversity of languages, differences of corresponding cultures, problems of interpersonal communication, changes in literary fashion, and distinct kinds of content and the circumstances in which translations are to be used.

Tytler [5] also introduced three laws of translation: (1) the translation should give a complete transcript of the ideas of the source language text; (2) the style of manner of the target text should be of the same character with that of the original one; and (3) the translation should have all the ease of the original text.

2.2 Machine translation and Google Translate

2.2.1 Machine translation

Alcina [6] defined machine translation (MT), also named as Automated Translation, as “a sub-field of computational linguistics that investigates the use of computer software to translate text or speech from one natural language to another”.

It dated back 1933 when P.P. Telojamsky, a Russian scientist, proposed using machine in translation. In 1954, the first MT system was born and successfully translated a Russian material into English. In 1976, MT technology was developed by Canadian Bureau of Translation, using TAUM-METEO translation system to translate weather report.

Translation versions using MT are automatically translated from one language to one or more other languages, without human intervention in the translation process.

According to Lin et al. [7], MT has been an important research topic for several decades in the field of artificial intelligence. More MT services have been provided by companies, and custom domain names or professions have been available in translation software. This helps improve the quality of translation results. However, the accuracy of documents has been far from the high requirement of qualities. Human intervention is still an alternative way to improve the quality of translation products.

2.2.2 Google Translate

Google Translate (GT) is a machine translation tool developed by Google. First developed in 2007, using a system named SYSTRAN, and then using the principle of Statistical Machine Translation in 2010 when being moved to a new platform of machine translation, GT has improved the quality of translation texts over time as more and more texts are loaded with diverse structures and contexts, creating a huge source of corpus.

With the adoption of Neural Machine Translation technology for 110 language pairs since 2016, GT has produced more accurate translation versions than other machine translation applications. It has translated texts at a sentential level instead of at a basic level, an independent substitution of words or phrasal level into broader contexts, increasing levels of translation accuracy.

Recently, GT has become the most popular machine translation tool since it offers both a website interface and mobile applications that users can use in their Android and IOS operating cellphones to get texts translated from a source language to a target language. In addition, GT has supported more than 100 languages at different levels, including Vietnamese. The application facilitates speech recognition, translating entire webpages, and uploading entire text files to speed up translation. Also, since GT is free for users to access; over 5 hundred million people all over the world use GT for their translation purpose.

2.3 Relevant studies on using Google Translate in translation

It cannot be denied that the application of GT in translation is increasingly popular although the accuracy of translation products has still been in questioned by many researchers. A lot of studies on using GT in translation and translation teaching and learning as well as the accuracy levels of texts translated by GT have been done.

Nguyen et al. [1] conducted the study to assess the quality of a text translated from English to Vietnamese using GT, compared with the human-translated translation among students majoring Management Accounting in English in Hanoi University of Industry. The research shows that Google Translate is the most successful in the word level and the least useful in grammar because accuracy is often reduced in some complex cases. Therefore, automatic translation is still not a substitute for translators and still requires human intervention and correction.

Zafitri [3] also carried out his research on Mathematic students using GT in learning English and concluded that Google Translate has a high effectiveness in the process of translation despite its drawbacks.

Another research was done by Aslerasouli et al. [8] on undergraduate students of English Translation translating texts in Physics and Politics using GT. The authors compared both human translations and machine translations to investigate their qualities. The findings indicated that there is a significant difference in the quality of human translations and machine translations in favor of human translations and the mode of translation affects its quality, but there is no statistically significant relationship between translation errors and translation modes.

Also, the study was done by Anggaira [9] on high school students’ translations from English to Indonesian to determine and then attempt to analyze the aspects of language errors that appear on the machine translator from Google Translate. The results indicate that GT translates word by word, but overlooks sentence context, resulting in morphological, syntactic, and semantic errors. Therefore, human improvements to the text of the translation are required to get translation accuracy.

The same results are found in the study conducted by Dwinanda [10] on the effectiveness of using GT in translation. The research concludes that GT can be successful if it translates at a basic level, word level, but GT fails to translate at a sentential level. Therefore, Google Translate is only a pre-translation that still needs a lot to be edited.

It can be noted that these previous researches used GT in the pre-translation stage and human intervention in the postediting stage to better the quality of the translation products although there were some differences in the participants (English major students or English nonmajor students, high school students or undergraduates), the language used in the source texts (the source text were written in English or in the participants’ mother tongue), and the purposes of the studies.

The author of this article, on the other hand, wished to adopt GT in the post-translation stage after analyzing and paraphrasing Vietnamese source texts in the pre-translation stage to minimize morphological, semantic, and syntactic errors in the target texts.

2.4 Some characteristics of Vietnamese compared to English

According to Pham [11], Vietnamese is an isolating language, so it does not use bound morphemes to express grammatical features such as number (singular/plural) and tense but relies on word order and function words.

In terms of word, there have been controversial ideas of what constitutes a word in Vietnamese. When viewing that a syllable bearing a meaningful unit is a word, “me con” can be orthographically separated into 2 words “me” and “con,” denoting two different subjects and translated into two English words “mother” and “child”. Also, “me con” can be orthographically separated into 2 words “me” and “con,” denoting a single subject and translated into two English words “my mother”. Moreover, “me con” may be viewed as a compound word and translated into mother–child relations because it signifies a single concept of mother–child relations (Do [12]) (illustrated in Table 1).

Vietnamese words	Lexical-semantic meaning	English translation
me con	Be orthographically separated syllables and denotes two different subjects	“Mother” “child”
	Be orthographically separated syllables but denotes a single subject	My mother
	One compound word, signifies a single concept of mother–child relations	Mother–child relations

Table 1.

Vietnamese word, its meaning, and English equivalences.

At the lexical–semantic level, words in Vietnamese and English share some similar characteristics as follows: (1) Words can be divided into content and function words. Content words bear lexical meaning, whereas function words relate content words to each other; (2) Content words may be further divided into word classes including nouns, verbs, and adjectives; and (3) Content words may have more than one meaning or belong to more than one-word class, with meaning and grammatical class disambiguated by sentence context.

However, there are some differences in English words and Vietnamese words when word class is changed (Bauer [13]). In English, words may keep the same form (e.g., tree bark vs. dogs bark) or change in form (e.g., sit in the chair vs. he chaired the meeting). In Vietnamese, there is no form change (Table 2). Therefore, word forms that may serve as nouns as well as verbs can only be distinguished within the context of each sentence.

Vietnamese	English translation
Anh ấy quyết định (verb) hủy bỏ cuộc họp.	He decided (verb) to cancel the meeting.
Quyết định (noun) của anh thật sáng suốt.	His decision (noun) was wise.

Table 2.

Vietnamese and English words at the lexical–semantic level.

Regarding Vietnamese pronouns, Pham [11], Tang [14], Luong [15], and Nguyen [16] have the similar idea that most Vietnamese kinship terms may be used as pronouns to reflect age, status, gender, and blood relations. Kinship terms that serve as pronouns are used with persons within and outside of one’s family. Within the family pronominal, kinship terms distinguish between paternal and maternal sides of the family, age, gender, and blood relations as opposed to in-law status. Unfamiliar speakers and listeners also refer to each other and themselves differently depending on social factors, including age and status (Table 3).

Vietnamese words	Age	Status	Gender	Blood relations
chú	Older than the speaker in most situations	Has higher status than the speaker in some situations	Male	Paternal side
cô	Older than the speaker in most situations	Has higher status than the speaker in some situations	Female	Paternal side

Table 3.

Vietnamese kinship terms as pronouns.

Moreover, the concepts of number (singularity or plurality) and person (speaker, listener, or third party) do not exist in Vietnamese pronouns, so a quantifier is added before a pronoun in order to indicate plurality in Vietnamese, and the meaning of the person reference can only be interpreted within the sentence or paragraph context. These language characteristics of Vietnamese are not found in English, which leads to different equivalences found in English translation depending on different contexts (Table 4).

Vietnamese kinship terms	English equivalences
chú (singular) (e.g, Chú đi đâu đấy?)	You (Where do you go?)
các chú (plural) (e.g, Các chú đi đâu đấy?)	You (Where do you go?)
cô (singular) (e.g, Cô đi đâu đấy?)	You (Where do you go?)
các cô (plural) (e.g, Các cô đi đâu đấy?)	You (Where do you go?)
cháu (singular) (e.g, Cháu đi đâu đấy?)	You (Where do you go?)
các cháu (plural) (e.g, Các cháu đi đâu đấy?)	You (Where do you go?)
bác (singular) (e.g, Bác đi đâu đấy?)	You (Where do you go?)
các bác (plural) (e.g, Các bác đi đâu đấy?)	You (Where do you go?)
anh (singular) (e.g, Anh đi đâu đấy?)	You (Where do you go?)
các anh (plural) (e.g, Các anh đi đâu đấy?)	You (Where do you go?)
chị (singular) (e.g, Chị đi đâu đấy?)	You (Where do you go?)
các chị (plural) (e.g, Các chị đi đâu đấy?)	You (Where do you go?)

Table 4.

Vietnamese kinship terms and English equivalences.

With regard to phrases and sentences, Vietnamese phrases and sentences are structured through two ways of combining words and using function words, which play a very important role in Vietnamese grammar. Due to the feature of a language with no bound morphemes, Vietnamese verbs do not have morphemes to express grammatical features of time (present, past, future), tense, and aspects (progressive, perfect) as in English verbs (Vo [17]). Therefore, function words are usually put in front of main verbs to express these grammatical meanings (Table 5).

Ways of structuring Vietnamese phrases and sentences	Vietnamese	English translation
Combining words	Anh ta lại đến.	He came again
	Lại đến anh ta.	It’s his turn again.
Using function words (của, và, vì)	anh của em	My older brother
	anh và em	You and I
	Anh vì em.	I do it for you

Table 5.

Ways of structuring Vietnamese phrases and sentences.

It was hypothesized that these characteristics of Vietnamese may pose a challenge to corpora data and caus errors if GT is used as the first step in translating a text from Vietnamese to English. Hence, in the research, Vietnamese source text analysis and paraphrasing are supposed to be done before GT application in order to produce high-quality translation versions from Vietnamese to English.

3. Research methodology

3.1 Research context and participants

The study was conducted on a class of 38 juniors studying at Van Lang University in Ho Chi Minh City, Vietnam, during the second semester of the 2022–2023 academic year. The study participants majoring in Translation and Interpreting in English were divided into seven groups to deal with their translation assignments. The course they were taking was Translation 2, which lasts 12 weeks. Their weekly assignments are Vietnamese-English translation and vice versa, but only Vietnamese-English translation was the focus in the study.

3.2 Methodology and research instruments

Since the participants’ perspectives and experiences were central, qualitative/interpretative method was deemed appropriate for the study. The study made use of nonparticipant observations and semi-structured interviews in the duration of the course. Nonparticipant observations were used as an additional data tool for the present study because this research instrument allowed understanding the phenomenon in context while still being separated from what was being observed. Observations helped contextualize findings from interviews and ensured that the data from interviews were reliable. Semi-structured interview was adopted as it allowed for in-depth responses rather than yes or no answers.

In our study, on the second week of the course, a demonstration of translation using GT was given to the students and a Vietnamese text was used as a sample. A weekly assignment was then given to the groups of students, and their translation activities were observed by the author. The author’s observations were done in 10 weeks. Every week, a group’s final product would be randomly chosen to get comments from other class members as well as the lecturer, and then, an interview was given to the group.

The translation process including 2 stages was demonstrated in the study as follows:

1. Pre-translation stage: Analyzing and paraphrasing the source text

   Step 1: Read through the text and get the main idea of the text.

   Step 2: Identify the time setting and the aspect in the text to identify tenses that are likely to be used in the target translation text.

   Step 3: Separate every Vietnamese sentence into different syntactic units such as subject, verb, object, adverb, clauses, and phrases.

   Step 4: Paraphrase these sentences simulating the structure of English sentences.

2. Translation stage: Using GT and editing the target text with human edition

   Step 5: Use GT to translate the paraphrased sentences.

   Step 6: Edit the translation text to have the final product.

4. Results and discussion

4.1 What difficulties may learners have when translating a text from Vietnamese to English?

The results from the study indicate that the biggest difficulties most of the groups dealt with were the structure of sentences, particularly with long and complex sentences. Some of their ideas were “it was impossible for us to identify which is the main clause and which is the subordinate clause,” or “the sentence is long and we found it difficult to locate its subject, its verbs and other elements,” or “it took much time to identify the subject and its verb in a sentence”. They also said that “Vietnamese grammar is confusing; it is not easy to identify a phrase is a subject or a predicate”. In addition, they said, “Subjects in Vietnamese texts are often very long, so it is not easy to locate them” or “Google Translate works best with simple and clear sentences. Avoid using complicated vocabulary or grammar structures that may confuse the translation tool”.

The students’ answers also show that vocabulary, particularly nouns and verbs, was also a challenge to the students. Some students found troubled to deal with compound nouns and noun phrases or verb phrases: “We could not identify which word is the main noun and which one functions as its modifying elements in a noun phrase, and which word is the main verb and which one functions as its modifying elements in a verb phrase.” Besides, they could not be sure if a word in a Vietnamese sentence was a verb or a noun when looking at the surface structure of the sentence. In addition, in some cases, they did not understand the meaning of some Sino-Vietnamese words which are part of Vietnamese language (e.g, đại lý bao tiêu, a technical term means offtake agency).

Regarding vocabulary use, the term “the contexts of the source text” is also mentioned when the seven groups considered choosing appropriate vocabulary. “Google Translate is unable to understand the contexts of the source text or the target text, which makes it highly possible that Google Translate sometimes offers nonsensical literal translations;” “I replace some of the wrong word choices with the more appropriate terms when they are needed in the different contexts;” “Google Translate may not always consider the context of the text being translated. Keep in mind that certain words or phrases may have different meanings depending on the context in which they are used” or “Certain phrases can also get lost in translation without the right context”.

It is surprising that there were two ideas that cultural factors implied in Vietnamese texts, and confusing meaning in Vietnamese texts also created some challenges to their translation. “You must carefully read the text, especially the specialized terms, slang words, to make sure that you can understand the layers of hidden meanings” or “Google Translate cannot recognize idioms, slangs”.

4.2 How can GT be used effectively and efficiently to produce high-quality translation versions from Vietnamese into English?

Table 6 displays a source text in Vietnamese and a translation version from GT without any modification. The two sentences in the Vietnamese source text are translated into three English sentences using GT in which Sentence 1 is not grammatically correct because it does not have the subject and Sentence 2 and Sentence 3 are also grammatically and semantically incorrect.

Source text written in Vietnamese

Translation version into English using GT without any modification

Sentence 1: Thực hiện Quyết định số 53/1999/QĐ-TTg của Thủ Tướng về một số biện pháp khuyến khích đầu tư trực tiếp nước ngoài, từ ngày 1.7.1999, giảm giá một số mặt hàng, phí, lệ phí một số loại dịch vụ cho các doanh nghiệp có vốn đầu tư trực tiếp nước ngoài. Sentence 2:Theo đó, các doanh nghiệp có vốn đầu tư nước ngoài được hưởng giá mua điện trong giờ sản xuất bình thường 7,5 cents/KWH, giá lắp đặt điện thoại nội hạt đối với doanh nghiệp và người nước ngoài áp dụng như mức quy định đối với doanh nghiệp trong nước và người Việt Nam, giá cước thuê bao điện thoại là 10USD/máy/tháng và giảm bình quân giá cước viễn thông quốc tế hiện hành từ Việt Nam đi các nước.

Sentence 1: Implementing the Prime Minister’s Decision No. 53/1999/QD-TTg on A number of measures to encourage foreign direct investment, from July 1, 1999, reducing prices of a number of goods, fees and charges for a number of services for foreign direct investment enterprises.; Sentence 2: Accordingly, foreign-invested enterprises are entitled to the electricity purchase price during normal production hours of 7.5 cents/KWH, the local telephone installation price for businesses and foreigners is applied as regulated rate.; Sentence 3: For domestic enterprises and Vietnamese people, the telephone subscription fee is 10 USD/phone/month and reduces the average current international telecommunications charges from Vietnam to other countries.

Table 6.

Translation version into English using GT without any modification.

Table 7 depicts the pre-translation stage in which the source text is analyzed and paraphrased before the paraphrased source text is translated using GT. The two sentences in the Vietnamese source text were paraphrased; the word order was adjusted; some words were added; and some words were omitted or replaced.

Source text in Vietnamese	Analyzing the source text	The source text paraphrased
Sentence 1: Thực hiện Quyết định số 53/1999/QĐ-TTg của Thủ Tướng về một số biện pháp khuyến khích đầu tư trực tiếp nước ngoài, từ ngày 1.7.1999, giảm giá một số mặt hàng, phí, lệ phí một số loại dịch vụ cho các doanh nghiệp có vốn đầu tư trực tiếp nước ngoài.	Context time: future aspect: not perfect tense: simple future Sentence structure: prepositional phrase + Subject + Verb prepositional phrase: “Thực hiện … nước ngoài, từ ngày 1.7.1999,”: “Thực hiện” (verb) is obmitted, “Theo” (function word) is added subject: “giá một số mặt hàng, phí, và lệ phí một số loại dịch vụ cho các doanh nghiệp có vốn đầu tư trực tiếp nước ngoài” verb: giảm (giá)	Theo Quyết định số 53/1999/QĐ-TTg của Thủ Tướng về một số biện pháp khuyến khích đầu tư trực tiếp nước ngoài, từ ngày 1.7.1999, giá một số mặt hàng, phí, và lệ phí một số loại dịch vụ cho các doanh nghiệp có vốn đầu tư trực tiếp nước ngoài sẽ được giảm
Theo đó, các doanh nghiệp có vốn đầu tư nước ngoài được hưởng giá mua điện trong giờ sản xuất bình thường 7,5 cents/KWH, giá lắp đặt điện thoại nội hạt đối với doanh nghiệp và người nước ngoài áp dụng như mức quy định đối với doanh nghiệp trong nước và người Việt Nam, giá cước thuê bao điện thoại là 10USD/máy/tháng và giảm bình quân giá cước viễn thông quốc tế hiện hành từ Việt Nam đi các nước.	Context time: future aspect: not perfect tense: simple future Sentence structure: adverb + Clause 1+ Clause 2 + Clause 3 adverb: “Theo đó,” Clause 1: “các doanh nghiệp có vốn đầu tư nước ngoài được hưởng giá mua điện trong giờ sản xuất bình thường 7,5 cents/KWH” Clause 2: “giá lắp đặt điện thoại nội hạt đối với doanh nghiệp và người nước ngoài áp dụng như mức quy định đối với doanh nghiệp trong nước và người Việt Nam,” Clause 3: “giá cước thuê bao điện thoại là 10USD / máy / tháng” Clause 4: “và giảm bình quân giá cước viễn thông quốc tế hiện hành từ Việt Nam đi các nước.”	Theo đó, giá mua điện trong giờ sản xuất bình thường sẽ là 7,5 cents/KWH cho các doanh nghiệp có vốn đầu tư nước ngoài; giá lắp đặt điện thoại nội hạt cho doanh nghiệp nước ngoài và người nước ngoài sẽ bằng với giá lắp đặt điện thoại nội hạt cho doanh nghiệp trong nước và người Việt Nam; giá cước thuê bao điện thoại sẽ là 10USD/máy/tháng; và giá cước viễn thông quốc tế hiện hành từ Việt Nam đi các nước sẽ giảm bình quân.

Table 7.

Analyzing and paraphrasing the source text.

Translation stage is indicated in Table 8. The source texts paraphrased are then translated using GT. In the translation version using GT without human edition, verb tenses are correct; the sentences are grammatically and semantically right to some extent. However, human edition that is mainly related to word choice is certainly a necessary requirement to have the best quality translation product.

The source text paraphrased	Translation version using GT without human edition	Translation version using GT with human edition
Theo Quyết định số 53/1999/QĐ-TTg của Thủ Tướng về một số biện pháp khuyến khích đầu tư trực tiếp nước ngoài, từ ngày 1.7.1999, giá một số mặt hàng, phí, và lệ phí một số loại dịch vụ cho các doanh nghiệp có vốn đầu tư trực tiếp nước ngoài sẽ được giảm.	According to the Prime Minister’s Decision No. 53/1999/QD-TTg on a number of measures to encourage foreign direct investment, from July 1, 1999, the prices of some Goods, fees, and charges for a number of services for foreign direct investment enterprises will be reduced.	In accordance with Decision No. 53/1999/QD-TTg by the Prime Minister concerning a number of incentives to be granted to foreign direct investment, from July 1st 1999, the prices of a number of commodities, costs, and fees charged on several services offered to foreign -invested enterprises will be reduced.
Sentence 2: Theo đó, giá mua điện trong giờ sản xuất bình thường sẽ là 7,5 cents/KWH cho các doanh nghiệp có vốn đầu tư nước ngoài; giá lắp đặt điện thoại nội hạt cho doanh nghiệp nước ngoài và người nước ngoài sẽ bằng với giá lắp đặt điện thoại nội hạt cho doanh nghiệp trong nước và người Việt Nam; giá cước thuê bao điện thoại sẽ là 10USD/máy/tháng; và giá cước viễn thông quốc tế hiện hành từ Việt Nam đi các nước sẽ giảm bình quân.	Accordingly, the electricity purchase price during normal production hours will be 7.5 cents/KWH for foreign-invested enterprises; the installation price of local telephones for foreign enterprises and foreigners will be equal to the price of installing local telephones for domestic enterprises and Vietnamese people; phone subscription fee will be 10USD/phone/month; and current international telecommunications charges from Vietnam to other countries will decrease on average.	Accordingly, the power cost during the normal operating hours will be 7.5 cents/KWH for foreign-invested enterprises; the in-country telephone installation costs for foreign enterprises and foreigners will be the same as those for domestic enterprises and Vietnamese people; the phone subscription rate will be 10USD/phone/month; and the current international telecommunications rates from Vietnam to other countries will decrease on average.

Table 8.

Using GT and editing the target text.

5. Conclusion

Based on the results collected and analyzed, the following conclusions can be drawn:

1. The difficulties the learners may have when translating a text from Vietnamese to English are mainly divided into two groups including difficulties at grammatical level and vocabulary level. The term “grammatical level” refers context, sentence structure, and phrase structure of the source text, while the term “vocabulary level” contains words, technical terms, or terms concerning cultural aspects of the source text.

2. GT is used more effectively and efficiently producing high-quality translation versions when human intervention is made before GT application and human edition is done after GT adoption. In response to the difficulties at grammatical level, pre-translation stage with analysis and paraphrase of the source text should be done before GT is applied. Translation stage with GT application and human edition is done to deal with the difficulties at vocabulary level.

Besides, translation strategies and techniques are also practiced and learners’ translation ability is also improved, which is the important goal of translation teaching and learning.

6. Recommendations and implications

While analyzing and paraphrasing the source text before GT application allowed learners to have more opportunities to practice translation techniques and to produce good translation products in this study, it is recommended that teachers’ demonstration at the very beginning of the course and teachers’ feedback after students’ translation are extremely important. The demonstration provided is a vital guidance of translation techniques that students should master to do their tasks efficiently and effectively; the feedback given may be extensive explanation of specialized notions, or knowledge of vocabulary and grammatical structures, which learners need.

Moreover, learners’ competence in the source and target languages also plays an important role in the success of the machine translation application as it is required that potential translators should rely on a wealth of vocabulary, grammar, and cultural aspects of these languages in order to do text analysis and paraphrase. To do so, teachers’ choice on the source text and teachers’ feedback as a grammar review after each translation product as are extremely crucial, which may contribute to learners’ progress.