Do you plan to align all your legacy translations to create a corpus or a translation memory, but have no time to make it?

You are not alone in this. There are more and more needs of creating a database or corpus from the legacy files, but you often end up with disappointment. If you are multilingual translation company, the problem will be more complex. Your PMs do not know for sure if one segment to another segment matches due to the language barrier. They will get confused when the alignment involves big file or data. Thus, this file alignment should be done by a native translator, meaning that the aligner (we refer to the person) should understand the source and the target well. The result will be no segment is misaligned and your corpus or translation memory can be used as valid reference to handle next projects.

What is alignment actually? It is a process of aligning two legacy documents (in any accessible formats, e.g. .xls .doc, .xlsx .docx, etc.) to be reused for translation memory or corpus, normally the result of aligning can be in the form of .TMX files. Thanks to this process, you do not have to create the TM or corpus or language database from the scratch. You just need to extract some legacy translations to match the source and you can then use it for your translation process. It will save cost in case some matches are found in a need-translation document.

The process seems easy, but if we looked closely, this process is not that easy - we often find misaligned segment; this could be the first and foremost issue we will encounter when aligning. Being able to handle this issue means that you are 80% done with your alignment job! Inside the process, we have to:

• join the misaligned segments to match the target or source, or
• split segments if they are not in relevant segments
• check the whole aligned files on a segment basis
• confirm the aligned segments, and
• export the aligned files to the desired formats.

That process can take either longer or shorter time depending on the translation style as well. If the translation is made on a free approach basis (noting no punctuation), the process can be very long and daunting. Otherwise, the process can take shorter time; the translation structure is made exactly based on the source rules.

Aligning the documents, we often read our client's request first before we decide the tools to use. We normally use:

• memoQ
• SDL Trados
• Regular Word Processors (Excel, Word).

We can use client's tool if available but it is very rare. If you are in need of this alignment service for English to Indonesian or Indonesian to English to be used as new TM or corpus or database, feel free to contact us. We can help you with the alignment. The price and turnaround? Let's discuss!