The Difference Between Machine Translation Vs Computer assisted Translation (CAT)

The Difference Between Machine Translation Vs Computer assisted Translation (CAT).

There is a significant difference between the two, with very different results. The terms “computer-assisted translation” and “machine translation” sound similar, and it’s easy to get them confused.

Machine Translation: Fast and Cheap, but Inaccurate

   Machine translation is accomplished by feeding a text to a computer algorithm that translates it automatically into another language. That is, no human is involved in the translation process.

   The advantages of machine translation include cost and speed. Computers can process a machine translation almost instantly. There are free programs such as Google Translate that can translate relatively short texts instantly, but if you need to translate a very long document, you can purchase software that can process an unlimited amount of text at the cost of the software alone. There is also software available that can be integrated with other computer and online tools, providing instant translations in various contexts.

    The major disadvantage is lack of accuracy. If you’ve ever used Google Translate to attempt to understand a text in a foreign language, you will know that this method does not produce a particularly natural-sounding or accurate translation. Language is highly complex and dynamic, and while this type of translation technology has improved greatly over the years, it will never be able to completely accurately identify the nuances of each language and transfer them into another language.

   It is possible to hire a “post-translation editor” to look over the translation and correct errors, but it can be harder to correctly deduct the meaning of a sentence from its machine translation than from its original language. Translators hired to “smooth out” such translations sometimes end up asking clients to send them the original text because the translation was unintelligible. This is a big waste of everybody’s time!

   The best use for machine translation, then, is when you need to understand the general gist of a text. If you need an accurate translation that anyone can understand, you’ll want to opt for a computer-assisted translation.

Computer-Assisted Translation(CAT): Human Translation Enhanced with Computerized Tools

   Computer-Assisted Translation is a human translation carried out with the aid of computerized tools. That is, a human translator is the one reading and deducing the meaning of the source text and transferring it into the target language. They are simply utilizing computerized translation tools to help them work more quickly and accurately.

   You probably already use some of these tools yourself. For example, nearly every word processor, and many web browsers, have a built-in spell checker and/or automatic spelling correction function. This saves writers and translators a lot of time looking up words in the dictionary!

   Speaking of dictionaries, when a translator does need to look up a word, they can save time by using a computerized dictionary. As a translator, my most often-used tools are the multi-language dictionary (to help recall words that may be escaping me at that moment) and the thesaurus (to help me choose exactly the right word for my translation).

   More complex computerized translation tools include translation memory tools (databases of texts in multiple languages), terminology managers (that help translators maintain consistent terminology throughout the translation), terminology databases (to help translators locate the correct terminology for that field), bitext aligners (which align the source text and the translation for side-by-side comparison), and more.

Types of Computer-Assisted Translation(CAT):

1. TRANSLATION MEMORY SOFTWARE

    Translation memory software is the most well-known CAT tool. It divides the texts to be translated into units called “segments”. As the translator advances in the translation of the document, the software stores the text in a database of already translated segments. When the software recognizes that a new segment is similar to a segment already translated, it suggests that the translator reuse it. Some translation memory programs do not work with databases created during a translation, but with preloaded reference documents.

Some examples of translation memory software: Trados Workbench, DéjàVuX, SDLX, Star Transit, MultiTrans, Similis, MetaTexis.

2. LANGUAGE SEARCH-ENGINE SOFTWARE

    Linguistic search engines work like traditional search engines, except that they do not seek results on the Internet, but in a large database of translation memory. The goal is to find, in these banks, fragments of previously translated texts that match the new text to be translated. Linguee, a multilingual context dictionary, is one of them.

3. TERMINOLOGY MANAGEMENT SOFTWARE

     Among CAT tools, there is also terminology management software. With programs of this type, the translator has the ability to automatically search for the terms in a new document in a database. Some of these systems allow the translator to add, in the database, new pairs of words that match and verify text using various functions: the translator can then check whether this or that term has been translated correctly and consistently throughout the whole draft. Here are three examples of this type of software: SDL MultiTerm, LogiTerm and Termex.

4. ALIGNMENT SOFTWARE

    Text alignment programs allow the translator to build a translation memory using the source and destination of the same text: the software divides the two texts into segments and attempts to determine which segments agree with each other. The result of this operation can be imported into a translation memory software for future translations. Here are four examples of alignment software: Bitext2, Tmx Bligner, YouAlign and LF Aligner.

5. INTERACTIVE MACHINE TRANSLATION

    Automatic interactive translation resembles the programs you use on your cell phone for writing messages: the program tries to predict how the human translator would translate a phrase or sentence fragment.

OTHER LANGUAGE PROGRAMS OF HELP TO THE TRANSLATOR

Finally, you should also consider other very useful linguistic software for translators:

Spell checkers (Proofread).

Grammar checkers (Grammarly, Reverso)

Terminology databases or online dictionaries, such as TERMIUM Plus, and the IATE.

Search tools for “full text” and indexing which allow searches to be carried out into already-translated texts or reference documents of all kinds, such as for example, ISYS Search Software and dtSearch Desktop.

Concordant or matching software which are reference tools used to look up a word together with its context, whether in a monolingual, bilingual or multilingual body (such as a bitext or a translation memory).

Project management software. With this program, a project manager at a translation company can organize complex projects by assigning translation tasks to different translators and track the progress of each one.

Types of Machine Translation:

1.     Rule-Based Machine Translation (RBMT)

RBMT, developed several decades ago, was the first practical approach to machine translation. It works by parsing a source sentence to identify words and analyze its structure, and then converting it into the target language based on a manually determined set of rules encoded by linguistic experts. The rules attempt to define correspondences between the structure of the source language and that of the target language.

The advantage of RBMT is that a good engine can translate a wide range of texts without the need for large bilingual corpora, as in statistical machine translation. However, the development of an RBMT system is time-consuming and labor-intensive and may take several years for one language pair. Additionally, human-encoded rules are unable to cover all possible linguistic phenomena and conflicts between existing rules may lead to poor translation quality when facing real-life texts. For example, RBMT engines don’t deal well with slang or metaphorical texts. For this reason, rule-based translation has largely been replaced by statistical machine translation or hybrid systems, though it remains useful for less common language pairs where there are not enough corpora to train an SMT engine.

2.    Statistical Machine Translation (SMT)

SMT works by training the translation engine with a very large volume of bilingual (source texts and their translations) and monolingual corpora. The system looks for statistical correlations between source texts and translations, both for entire segments and for shorter phrases within each segment, building a so-called translation model. It then generates confidence scores for how likely it is that a given source text will map to a translation. The translation engine itself has no notion of rules or grammar. SMT is the core of systems used by Google Translate and Bing Translator, and is the most common form of MT in use today.

The key advantage of statistical machine translation is that it eliminates the need to handcraft a translation engine for each language pair and create linguistic rule sets, as is the case with RBMT. With a large enough collection of texts, you can train a generic translation engine for any language pair and even for a particular industry or domain of expertise. With large and suitable training corpora, SMT usually translates well enough for comprehension. The main disadvantage of statistical machine translation is that it requires very large and well-organized bilingual corpora for each language pair. SMT engines fail when presented with texts that are not similar to material in the training corpora. For example, a translation engine that was trained using technical texts will have a difficult time translating texts written in casual style. Therefore, it is important to train the engine with texts that are similar to the material that will be translated.

Source:

https://www.ulatus.com/translation-blog/machine-translation-vs-computer-assisted-translation-whats-the-difference/

https://www.onehourtranslation.com/translation/blog/cat-vs-machine-translation

Present Perfect Tense

DEFINISI DARI PRESENT PERFECT TENSE

Present perfect digunakan untuk menunjukkan hubungan antara masa kini dan masa lalu. Waktu tindakan atau kejadiannya adalah sebelum sekarang, tetapi tidak spesifik, dan kita sering kali lebih tertarik dengan hasil atau akibatnya daripada tindakan itu sendiri.

HATI-HATI! Mungkin ada tense dalam bahasa Anda yang memiliki bentuk yang sama, tetapi artinya mungkin TIDAK sama.

PRESENT PERFECT DIGUNAKAN UNTUK MENDESKRIPSIKAN:

• Tindakan atau situasi yang dimulai di masa lalu dan masih berlanjut hingga kini. I have lived in Bristol since 1984 (= dan saya masih tinggal di sana.)
• Tindakan yang dilakukan selama periode waktu yang belum selesai atau berakhir. She has been to the cinema twice this week (= dan minggu ini belum berakhir.)
• Tindakan berulang dalam periode yang tidak ditentukan antara masa lalu dan kini. We have visited Portugal several times.
• Tindakan yang baru saja selesai atau berakhir, dinyatakan dengan just. I have just finished my work.
• Tindakan di mana waktu tidaklah penting. He has read 'War and Peace'. (= hasil atau akibat dari tindakan membaca itu penting)

Catatan: Saat kita ingin memberikan atau meminta detail mengenai kapan, di mana, dan siapa, kita menggunakan simple past. Bacalah selengkapnya tentang cara memilih antara present perfect dan simple past tense.

TINDAKAN YANG DIMULAI DI MASA LALU DAN MASIH BERLANJUT HINGGA KINI

• They haven't lived here for years.
• She has worked in the bank for five years.
• We have had the same car for ten years.
• Have you played the piano since you were a child?

SAAT PERIODE WAKTU TINDAKAN BELUM SELESAI ATAU BERAKHIR

• I have worked hard this week.
• It has rained a lot this year.
• We haven't seen her today.

TINDAKAN BERULANG DALAM PERIODE WAKTU YANG TIDAK DITENTUKAN ANTARA MASA LALU DAN KINI

• They have seen that film six times
• It has happened several times already.
• She has visited them frequently.
• We have eaten at that restaurant many times.

TINDAKAN YANG BARU SAJA SELESAI ATAU BERAKHIR (+ JUST)

• Have you just finished work?
• I have just eaten.
• We have just seen her.
• Has he just left?

SAAT WAKTU TINDAKAN YANG TEPAT TIDAK PENTING ATAU TIDAK DIKETAHUI

• Someone has eaten my soup!
• Have you seen 'Gone with the Wind'?
• She's studied Japanese, Russian, and English.

Bacalah selengkapnya tentang cara menggunakan present perfect dengan kata ever, never, already, dan yet dan tentang cara menggunakan present perfect dengan kata for dan since.

MEMBENTUK PRESENT PERFECT

Present perfect dari kata kerja apa pun terdiri dari dua unsur: bentuk auxiliary verb to have (present tense) yang sesuai, ditambah past participle dari kata kerja utama. Bentuk past participle dari kata kerja beraturan adalah kata dasar + ed, seperti played, arrived, looked. Untuk kata kerja tidak beraturan, lihatlah tabel kata kerja tidak beraturan di bagian verb.

Positif
Subjek	to have	past participle
She	has	visited.
Negatif
Subjek	to have + not	past participle
She	has not (hasn't)	visited.
Pertanyaan
to have	subjek	past participle
Has	she	visited?
Pertanyaan negatif
to have + not	subjek	past participle
Hasn't	she	visited?

TO WALK, PRESENT PERFECT

Positif	Negatif	Pertanyaan
I have walked	I haven't walked	Have I walked?
You have walked	You haven't walked.	Have you walked?
He, she, it has walked	He, she, hasn't walked	Has he, she, it walked?
We have walked	We haven't walked	Have we walked?
You have walked	You haven't walked	Have you walked?
They have walked	They haven't walked	Have they walked?

Computer Assisted Language Learning

A Brief History of CALL

Computer Assisted Language Learning (CALL) is the general term for the range of processes and activities that employ computers in the teaching and learning of a new langauge.

In the history of CALL we can see the confluence of the latest technology as well as the most widely accepted language theories of the day.

The history of CALL is often divided into three phases:

1. Structural CALL
2. Communicative CALL
3. Integrative CALL

Starting in the ’50s and developing through the ’70s, we have what’s called Structural/Behaviorist CALL by Warschauer. This marked the era of Stimulus and Response. The computer prompts the student with a question (stimulus) and the student gives an answer (response) by filling in the blanks or choosing from a given set of choices.

The methods du jour were the Grammar-Translation and Audiolingualmethods. Language was seen as made up of discrete units, and these units were considered to be closely interconnected and interacting according to a predictable and explainable set of rules (grammar). Teachers taught the different rules of grammar and repetitively drilled their classes on different ways the rules can be correctly applied.

Computers at this stage were mainly utilized as devices that could present stimuli repetitively in exactly the same manner without ever getting tired. An example of this are the “listen-and-repeat” programs running in language labs at that time.

In the ’80s and ’90s came Communicative CALL. The Communicative Approach to language teaching came into being as a reaction to the Grammar-Translation and Audiolingual methods. This time, instead of teaching the language—its rules, syntax, phonemes and morphemes—teachers found ways to provide opportunities for students to actually use the language. They gave students tasks that can only be completed by using language. Communication and interaction were important.

And because such technology always comes in service of the language paradigm of the day, computers were used to reflect these ideas. Language drills were increasingly placed in the context of a communicative task—like programs that feature some cartoon character where students help him find his way home. Computer programs were designed to gauge comprehension with drills like paced reading and sentence reconstruction.

And developments in computer technology didn’t just affect the “testing” part of CALL. It really made teaching language more vivid. For example, the continued development in computer capabilities has resulted into crisper audio and video. So in addition to the drill formats, students can learn by watching videos of how native speakers actually interact. They can see how language is used in different situations, like in meeting a new person or asking for directions. Computers have given language learners a more vivid idea of what language is beyond the subject-verb agreements and the endless list of vocabulary words to be memorized.

The next phase of CALL is the Integrative Phase (which has reigned from 2000 onwards). First came the drills of the structural approach, then followed the skills in the communicative approach. Critics of the second phase say that the skills taught may be limited to the number and types of situations that may be presented to students. (We are not asking for directions or ordering food at the restaurant the whole time.)

There needs to be an integration of the (general language) knowledge presented in the first phase as well as the communicative skills of the second phase. So we have the integrative phase which blended the virtues of the previous decades into a technology that, for its part, has found its stride.

The development of the internet and hypermedia that can integrate, video and audio streaming, graphic-interactive content and virtual worlds, have redefined how learning is done. With today’s technology, you can develop speaking, listening, reading and writing skills concurrently and in the comfort of one’s private space and schedule.

The perfect example of this Integrative CALL is FluentUwith its interactive videos. You have a slew of different types of video content involving different types of topics, themes and situations (Phase 2). You have an interactive transcription where when you scroll over any word, out pops its own dictionary entry that explains various usage and rules of grammar and syntax for that specific word (Phase 1). With technologies like FluentU, you get the best of both worlds.

And that, briefly, is how CALL has developed over the years.

What technology can do to redefine the concepts of teaching and learning language will be up for grabs for the next game changer.

Advantages: The 2 I’s of CALL

Individualized

One of the advantages of CALL, in its present form, is the ability to cater to individual differences. Differences in learning styles, language skills desired, pacing and learning schedules can be easily accommodated.

It used to be that computer programs deliver a one-size-fits-all, cookie cutter material that can only be accessed after signing your university’s computer lab log book. Today, learning a language has not only been democratized, it has been individualized.

For example, you create a free account in any of the major language learning sites like Busuu and Babbel and you start your own learning journey without interference from others. There are no classmates, no group lectures and no chorus of students repeating after teacher.

You decide how much time you want to put in and when you want to access it. There’s no calendar for classes where you’ll be marked absent when you don’t show up.

You can access the materials anytime and anywhere you want. Actually, one way of looking at the history of CALL is by noticing how technology has individualized language learning. The university’s mainframes and language labs used to have a monopoly on some clunky software. Then came the PC in the 90s and were computers found a home in practically every home. Today, with mobile technology, language learning can be had on the go, while sitting on the bus, while waiting in line at the Apple store or even while taking a shower.

Interactive

CALL has come so far along that it can virtually replace an actual teacher asking the class, “So, what do you guys think? What do you want to do next?”

Well, not all teachers want input from their students. The advantage computers have is that they do need an input in order to run. That means they’re inherently interactive. Over the decades, the complexity of this interaction has been increasing. From the simple stimulus-response in early computers where students are practically passive learners, we now have CALL actually “learning” and “remembering” student preferences. From a simple text presentation, we now have gamified graphics like Mindsnacks.

The individualized nature of CALL has led to the second “I.” Interactive means that when you click on something, the computer responds. There’s enough flexibility built into the technology so that what happens in the lesson is largely up to you. Do you want to take it in this or that direction? Not only can the students choose which topics to study, skip or which ones to tackle first, they can click also forward and backward, and the computer obliges their commands.

The interactive nature of today’s CALL ensures that learning is always a two-way street. Students do have a say in what they want to learn. CALL is dynamic, not static. Robust not rigid.

How Is CALL Used?

To Teach

CALL applications can be used by teachers as technology partners in running their classrooms from the initial intro of language concepts to the giving of electronic homework. Students are using computers in practically every other aspect of their lives anyway, from locating the nearest coffee shop to shopping for new shoes. So why not throw learning a new language to the mix?

CALL, in addition to integrating technology in the learning process, also helps solve classic teacher problems like capturing student attention, maintaining student interest, holding focus and increasing engagement. Teachers can benefit from the great variety of interactive activities, games, songs and stories that make language learning not only painless but also fun.

Applications like the award-winning Language Nut was developed for this very purpose and looks to be a complete solution and curriculum partner for language teachers. (It was developed, after all, by former language teachers.) It supports four language skills—listening, reading, writing, speaking—and has an immersive interface that’s easily addictive. In the world of Language Nut, you sing songs, play games, listen to stories and remember vocabulary all the way to fluency.

To Reinforce

CALL can also be used to reinforce a teacher’s classroom lessons and activities. When educators need help in making lessons more vivid and when they need the concepts to come alive, instead of pasting cut-outs and visual aids on the board they can make use of multimedia lessons offered in CALL.

FluentU is one example of CALL that can be used in every phase of teaching language. Its concerts, interviews and music video clips, for instance, can scaffold linguistic discussions given by the teacher, providing a different look and a new approach to the lesson. They put new language in context and breathe life into it. Multimedia content can effectively substantiate topical lessons, from greetings and introductions to talking about the weather, food and even sports.

But CALL doesn’t only give students a clear line of sight (and sound) on what the teacher is talking about. It has capabilities beyond what any human can do. FluentU has interactive transcripts, which means practically everything you need to know about a specific word—like definitions, in-context usage and pronunciation—pops out the moment you roll the cursor over said word. The learn mode of this program employs SRS to introduce and reinforce new vocabulary, grammar patterns, expressions and even full sentences, and incorporates video clips into its flashcards and dynamic learning games.

CALL doesn’t have the physical limitations that cap humans. That’s why it can bridge the gap when human endurance and consistency need a boost. For example, a teacher can only repeat the lessons so many times. But repetition is key if no child in class is to be left behind. CALL apps, videos and programs can be run and rerun as many times as necessary, without fatigue and diminishing returns, and irrespective of geography or time. That means students can review and study the lessons long after the teacher has gone home and sound asleep.

To Test

There will probably never be a substitute for a teacher or a native speaker to determine whether a student has actually become fluent with the language, but CALL has become very good at assessing competency with subsets of a language. For example, it can easily determine if a student has mastered specific topics, like grammar and vocabulary.

But beyond simple testing really, CALL has been able to integrate both teaching and testing in a single stroke of a mouse. With programs like Duolingo, Memrise and Brainscape, there’s very little time gap between teaching and testing, or rather, very little difference between teaching and testing at all.

For example, in a simple translation exercise, the French word for smile (sourire) might be presented in a slide or flashcard with pics and an audio feed. With a simple click of the “next” button, a user might immediately be shown a slide that testing “What is French for smile?” This encourages the learner to recognize the word and produce the word in different contexts.

CALL is free of subjective biases and can faithfully follow a predetermined set of algorithms. That is, if a user shows mastery over certain topics or words then the program proceeds to other more difficult material. If they don’t have this knowledge ingrained yet, then it repeats the material until it has determined that the user has exhibited sufficient knowledge of the subject. In a way, the program tells the student, “Hey, you haven’t really learned this word yet, so I’m going to present it a couple more times so you can have it saved in your long-term memory.”

To Practice

CALL can be used even when classes are out and in the teacher’s absence. Language learning technology in its present form is student-initiated and student-centered, giving all the time and all the room in the world for students to practice. Language practice can be had in the privacy of one’s room and at a moment’s notice. And the kicker is that students get to do all this without fear of being negatively judged by others.

And of course, CALL practice is equal parts learning and fun as exemplified by Mindsnacks—a gamified approach to language learning. For example, it has a game called Dam Builder where you shift wooden logs around so that, in the end, you’re able to pair corresponding words/phrases.

But probably one of the most important contribution technology has to language learning is that it has given learners access to the native speakers. Technologies like italki and Skype took language learners from hammering at the language alone, to working with native speakers, tutors or teachers sitting in their own rooms half a world away. In the past, this kind of practice could only be had by flying across oceans.

CALL makes everything that much easier. From the teaching, reinforcing, testing and practicing, CALL presents itself as a capable and consistent partner to both teacher and student. It changed the way languages are being tamed.

But for all its virtues, there’s one thing that will always remain in the human province. Motivation. The zeal to learn a new language will always be alien to technology. Technology can’t manufacture drive out of thin air, for it’s fashioned into the inner recesses of the human spirit.

Source: https://www.fluentu.com/blog/educator/what-is-computer-assisted-language-learning/