This content has been archived. It may no longer be relevant
Before we get into the details, let us first define the types of machine translation the current market offers.
If you check Wikipedia on “Machine translation,” you will find an interesting article on what the experts consider Machine translation, which shouldn’t be confused with the computer-aided translation tools (Translation Memory) professional translators utilize.
The following Machine translation paradigms are available:
- Rules-based
- Statistical
- Example-based
- Hybrid MT
Depending on what you are trying to achieve with your translation, any of those could benefit you.
Google Translate, which is probably the most widely used and easiest to access, is currently based on the statistical translation method.
Getting a rough draft of foreign language documents
If you find yourself in front of a document written in a language you don’t speak or read, then a machine translation tool could come in handy and provide you with a quick overview on what the document is about.
Use it for Website translations – cautiously
Many smaller companies opt to use the plug-in Google Translate offers on their website. It is always good to conduct a simple test to see if it would be appropriate for what you’d like to get accomplished. Take your content and have it translated into another language; once you have this text, use it as base and have it translated back into your language. With this back-to-back translation, you can easily determine if the translation tool is something that will work for you. A word of caution: We found that if the source text exists on the web already translated in the particular target language, then the translation will be fairly good. However, if the source text is unique, then the resulting translations leave much to be desired.
Knowledge-based translation – Hybrid MT
This can be a great solution for some global organizations especially for their after-market and customer-service initiatives. A combination of knowledge-based and Hybrid MT is a custom-made machine translation created out of a company’s own translation memory, dictionaries, phrases, and as much knowledge base as is available. This is knowledgebase is exclusive for this organization and usually takes several years to build. But once the quality is there, it can be a great tool. Let me give you an example on how it could be used:
Imagine you have a product or machine used by customers around the world. Instead of having a full-force service team set up in each country where you sell your machine, you can utilize a multi-lingual database, allowing the customer to phrase the problem he is experiencing in his language online. This will then be translated into the language of your service team. They respond with the solution to the problem and the answer is translated back into the language of your customer. This process can be automated, providing 24/7 care even in remote areas.
Conclusion:
While there are occasions when machine translation is acceptable to use, they are limited and should be qualified. Keep these tips in mind:
- Machine translation can provide an overview on content in other languages
- It is best to use machine translation on content that you translate into a language you speak
- Check the quality of the translation by making a back-translation and test the result
- Don’t expect an accurate translation
- Don’t use it to translate something into a language you don’t speak and expect the reader in the other language to be able to use it
Tell us about your project and let’s discuss the most streamlined approach to reach your translation goals.
