Home≫Column list≫What is automatic translation technology? Differences from machine translation and recommended services
What is automatic translation technology? Differences from machine translation and recommended services
Advances in automatic translation technology and introduction of highly accurate translation sites!
Advances in AI have dramatically improved the technology and accuracy of machine translation and automatic translation. With the development of technology and the progress of globalization, the use of machine translation and automatic translation is increasing in business. Furthermore, in today's world where the Internet and smartphones are widespread, anyone can use a free site to instantly translate into languages around the world. Here, we will explain machine translation and automatic translation, which are becoming more familiar, and introduce popular translation sites.
LaughteySo, unlike general-purpose AI automatic translation services, we offer low-cost, high-performance "Translation Memory Form” is provided. It is a service that can be used permanently because it starts from training as a dedicated concierge for customers. If you develop a dedicated customer concierge, the longer you use it, the more cost performance you can demonstrate.
There is no limit on the number of characters, and the fee is set for each requested manuscript.Inquiry formPlease feel free to contact us if you are thinking of using it.
table of contents
1.It can also be used for translating specifications! What is automatic translation?
1-1.What is the difference between machine and automated translation?
1-2.History of machine translation
2.The accuracy of Chinese translation is also improved! Types of automatic translation sites
2-1.Introducing the automatic translation site!
It can also be used for translating specifications! What is automatic translation?
In recent years, not only translation work using human hands but also translation using computers has spread. There are also multiple free translation sites, and anyone can translate immediately. The quality is improving day by day, and it is now possible to obtain highly accurate translations even for free. New technologies such as translation sites will also be useful when translating specifications used in business. Here, we will explain what automatic translation is, the difference from machine translation, and approach the history of machine translation that has developed along with the progress of AI technology.
What is the difference between machine and automated translation?
Many people have an image that both machine translation and automatic translation are translations by computers. They are the same in terms of translation, but there are technical differences.
◇Machine translation
Machine translation is translation done by a computer without human intervention. In machine translation, a machine performs the work of converting text information into other languages using a huge database. Machine translation systems are roughly divided into rule-based translation, statistical-based translation, and neural machine translation.
Among them, neural machine translation, represented by Google Translate, has a network structure similar to that of human brain neurons, and can produce highly accurate and natural translations.
◇ Automatic translation
Automatic translation is primarily speech translation. It translates the input voice on the spot and outputs it as voice. Since automatic translation is highly immediate, it is suitable for real-time communication between people who speak different languages. Useful in travel and business interactions.
As a mechanism, for the voice information that you want to automatically translate, first perform voice recognition and then convert it to characters, and then translate the recognized characters in the same way as machine translation. In other words, automatic translation means that machine translation is always performed in the process of translating in real time.
Due to the development of network systems, opportunities to communicate with foreigners and use foreign languages have increased even in Japan. Demand for translation is expected to increase further as globalization progresses in the future. If you remember systems such as machine translation and automatic translation, it will be useful when you need translation work in business situations or privately.
History of machine translation
The history of machine translation is also the history of AI (artificial intelligence). We will explore the history of machine translation from the time when computers themselves were not widely used to the present day when machine translation can be easily used online.
In the early 1950s, attempts were made to make computers perform intelligent activities. In 1954, IBM and Georgetown University jointly developed a machine translation system. Successfully translated Russian into English.
However, there were still few words that could be translated, and it was far from being put to practical use.
In the 1980s, research progressed in making computers memorize specialized knowledge in specific fields as rules. Rule-based translation appeared around 1985. This is a system that teaches translation rules to a computer, analyzes the original text, and outputs a translated text that matches the registered rules. In the late 1980s, statistical machine translation was developed. Machines themselves are now able to perform statistical processing from huge amounts of data and select appropriate translations.
Around 1990, a workstation equipped with a machine translation system was released in Japan. Although it can be called the first machine translation boom in Japan, it was not highly evaluated because it was expensive and lacked accuracy. After that, inexpensive translation software for general households was sold, resulting in a second boom. However, this was also not practical enough for business use.
In the 2000s, research into deep learning progressed, and neural machine translation using this technology improved the accuracy of machine translation. Neural machine translation was put into practical use in 2015, and from 2016 Google and Microsoft started services using neural machine translation.
In today's world where AI technology has improved dramatically, machine translation is also used in business as a practical system. The results of the research efforts of our predecessors have led us to where we are today.
The accuracy of Chinese translation is also improved! Types of automatic translation sites
China, which has grown as an economic superpower, is expanding its business opportunities to the world. Of course, Japan is no exception, and there will be more and more situations where Chinese translation is required. Along with this, it is important to have translation technology that is highly accurate and can be used without problems in business. Here, we will introduce an automatic translation site that can also be used for Chinese translation, and the topical machine translation "DeepL".
Introducing the automatic translation site!
What are the features and usage scenarios of popular free translation sites?
◇Google Translate
It is the world's largest free translation site that supports over 100 languages. In addition to text input by voice recognition, it also supports translation of entire web pages and various documents such as Word, Excel, and PDF. In the app version, information technology is utilized in various ways, such as a real-time translation function that translates characters as soon as they are entered, camera input, conversation mode, etc.
Furthermore, in 2016, we introduced neural machine translation as a new technology to achieve highly accurate translation. Although the translation technology is high, it is recommended when the amount of the manuscript to be translated is small because the data to be translated must be registered one by one.
◇Weblio translation
This is a translation site for Japanese-speaking countries operated by Weblio, Inc. You can translate text and web pages. Languages are English, Chinese, and Korean, and it is also possible to output translations by voice. You can also use the retranslate function to convert the translation back to the original language, which is useful for checking accuracy.
However, many of the translated sentences are literal translations, which is somewhat unnatural as a colloquial sentence. Bilingual translations for each word and sentence are displayed, so it is more suitable for learners than for business use.
◇ Excite translation
This is a translation site that supports 30 languages around the world. By selecting the field, it is characterized by being able to translate with higher expertise and accuracy. Fields include science, engineering, art, etc. For example, science is further subdivided into mathematics, physics, astronomy, and so on.
The translation is established as a sentence, but it often lacks accuracy when it is a long sentence. It can be said that it is suitable for translating short sentences in specialized fields.
What is the topic of machine translation "DeepL"?
Since its release in 2017, "DeepL" has gained popularity with tremendous momentum, and is said to have over 500 million users in China and other countries around the world. A free translation service from DeepL based in Cologne, Germany. It supports Windows and Mac, and has an app version and a browser version. Support for Japanese started in 2020. It supports translation of many languages such as English, Chinese, French, and German.
◇ How to use the browser version
If you register the original text in the input field, the translation will be generated almost in real time. The number of characters can be up to 5,000 characters, and there is also a read-aloud function for translations.
It supports Word and PowerPoint, and you can also translate each file. It is basically free, but there are restrictions on the number of files and characters. If you have many files or volume, you can use the paid version.
◇ Reason why DeepL is chosen
DeepL's biggest selling point is its high translation quality. Machine translation can look unnatural to native speakers. DeepL is a technology that uses neural machine translation to make up for the unnaturalness and create a natural translation. DeepL translations can include colloquial and informal expressions, dialects, and nuances. It can be said that this is an epoch-making translation service that uses technology to overcome parts that could only be done manually.
However, even if the expression is close to natural, it is still not perfect. Please note that the translation may be unnatural depending on the original text.
In that respect, Laughtey's TMF (Translation Memory Form) service is specialized for companies. Also, as the database accumulates, the speed will increase, and the accuracy of the second translation will also increase. As the number of translations increases, the accuracy of the translation improves significantly, so the unit price can be lowered. Registration in the dictionary database, the amount of information increases when the same thing comes, and the correction is also up-to-date.
Automatic translation technology that can translate specifications and Chinese is evolving day by day!
As globalization progresses, business often needs translation. In particular, the demand for Chinese translation is increasing. In order to overcome language barriers and achieve smooth communication, it is important to make effective use of technologies such as automatic translation and machine translation.
However, you need to be careful with free translation sites, as there are restrictions on the number of characters and files, and the translation may be unnatural.
If you need high-quality translation of large volumes of specifications, manuals, etc., Laughtey can help. We provide translation with good cost performance.
About Laughtey's translation agency service
If you want to use a high-tech automatic translation service, Laughtey
company name
Rafty Co., Ltd.
CEO
Tetsuji Okumura
Established
May 2014
address
1-3-14 Hiroo, Ichikawa City, Chiba Prefecture 272-0146
TEL
080-3785-3773
Business content
project controller
IT service implementation consultant
RFI/RFP creation
SE and PM agency service
Implementation of Salesforce
Website creation
Web application introduction support
BPO (business outsourcing service)
Secondhand goods sales (antique dealer)
URLs