# Inconsistent words check

**URL:** <https://forum.xbench.net/t/inconsistent-words-check/1102>\
**Category:** General Discussion\
**Created:** [January 19, 2024, 2:41am UTC](https://forum.xbench.net/t/inconsistent-words-check/1102 "2024-01-19T02:41:42Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![takehiro](https://avatars.discourse-cdn.com/v4/letter/t/3ec8ea/32.png) [@takehiro](https://forum.xbench.net/u/takehiro)\
**Post date:** [January 19, 2024, 2:41am UTC](https://forum.xbench.net/t/inconsistent-words-check/1102/1 "2024-01-19T02:41:42Z")

</div>

I’d like to know how to check the inconsistent words, such as “コンピュータ” and “コンピューター”, If Xbench can support this. “コンピュータ” and “コンピューター” are Japanese words.

---

<div class="post-metadata">

**Author:** ![omartin](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.xbench.net/omartin/32/82_2.png) [@omartin](https://forum.xbench.net/u/omartin)\
**Post date:** [January 19, 2024, 2:17pm UTC](https://forum.xbench.net/t/inconsistent-words-check/1102/2 "2024-01-19T14:17:03Z")

</div>

If you translate into Japanese, just create an Excel file and enter source term in column A and Japanese term in column B. Then, add the xlsx file to Xbench as keyterms.

Make sure to uncheck **Match Whole Word in QA Check for Key Terms** at **Tools \> Settings \> Miscellaneous** tab. At the QA tab, enable **Key Terms Mismatch** at Content check group of the QA options pane.

---

<div class="post-metadata">

**Author:** ![takehiro](https://avatars.discourse-cdn.com/v4/letter/t/3ec8ea/32.png) [@takehiro](https://forum.xbench.net/u/takehiro)\
**Post date:** [January 20, 2024, 9:40am UTC](https://forum.xbench.net/t/inconsistent-words-check/1102/3 "2024-01-20T09:40:24Z")

</div>

Hi, thank you for your advice.  
I tried according to your advice but failed.

Please let me change the example, which is the case that the target language is English.  
There are several names “George” in the source language.  
And then the target translations associated with the same source language “George” have two kinds, which are “Jorge” and “Jourge”. “Jorge” is correct and “Jourge” is not correct.  
I want to know the way to check this kind of mistranslation.

I think I can add the key term file with “George” in the source and “Jorge” in the target.

Thank you in advance.

---

<div class="post-metadata">

**Author:** ![omartin](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.xbench.net/omartin/32/82_2.png) [@omartin](https://forum.xbench.net/u/omartin)\
**Post date:** [January 22, 2024, 8:33am UTC](https://forum.xbench.net/t/inconsistent-words-check/1102/4 "2024-01-22T08:33:59Z")

</div>

Try the following search:

Source: `"(<[A-Z][a-z]+>)=1.*@1"`  
Target: `-@1.*@1`

Search mode: Regular Expressions  
PowerSearch, Case-Sensitive: on

Please note that this search may be really slow. Try only in Ongoing Translation files.

---

<div class="post-metadata">

**Author:** ![takehiro](https://avatars.discourse-cdn.com/v4/letter/t/3ec8ea/32.png) [@takehiro](https://forum.xbench.net/u/takehiro)\
**Post date:** [January 25, 2024, 9:08pm UTC](https://forum.xbench.net/t/inconsistent-words-check/1102/5 "2024-01-25T21:08:17Z")

</div>

Hi omartin,  
Thank you very much.  
I will try it!