I want to know the trick to transliterating names in to Japanese.

I previously asked this question and it was suggested to look up famous names.

That worked until I came up on one of my friend's last name, which is Cobaxin.

I couldn't find anyone famous/popular with that last name.

Is there a set of rules to follow when transliterating names into Japanese Katakana?

I wouldn't like to bother you guys for every single name that I can't figure out :P


The origin of the name is Spanish so in Spanish it's pronounced ko-ba-kzeen.

However in English it's pronounced ko-ba-kzen.

I'm thinking it's somewhere in between コバクゼン or コバクゼィン. What do you guys think?

  • I wish it was that simple, but even when I try and look up other more popular names to check if I did it right sometimes way off. So it's not that easy... – Tek Jul 3 '14 at 14:32
  • Edited to give you guys more details on the pronunciation. – Tek Jul 3 '14 at 14:48

Usually, words are transcribed based on how they sound—or how it is imagined that they'd sound, since many words are borrowed into Japanese via print and aren't actually heard.

  1. A Spanish-speaking friend offers /ko'βaksin/ as an IPA transcription of Cobaxin. Based on that, I think you can straightforwardly transcribe the Spanish version of the name in Japanese as コバクシン, replacing the consonants with their closest equivalents and adding the epenthetic vowel /u/.

    The vowel /u/ is the most common epenthetic vowel in Japanese and is usually the default choice after consonants other than /t/ and /d/. After these two consonants, adding /o/ is most common.

    These vowels are added when you need to break up an illegal consonant cluster. Consider the English word strength:

    ストレングス /sutoreɴgusu/

    Here, English strength /strɛŋθ/ has the consonant cluster /str/, which is not possible in Japanese, so vowels are inserted to break it up, resulting in /sutor/. (However, since the English has no /g/ sound in most dialects, we can guess that this transcription was made from an imagined pronunciation based on spelling. This is often the case.) As you can see, /u/ is added in each case, except after /t/, where /o/ is inserted instead.

    Generally speaking, you can get pretty far by adding these two vowels whenever necessary. But giving a complete set of rules for transcription is probably beyond the scope of a single answer, I'm afraid; I'll focus on Cobaxin for the rest of this answer.

  2. You indicate that in English the final vowel of Cobaxin is more like /e/ than /i/. If that's correct, then I think you could write コバクセン instead.

  3. In your transcription, you've written z rather than s. If you think this is more accurate, then コバクジン or コバクゼン would be fine, again depending on the final vowel.

  4. If you intend to write /si/ or /zi/, I would stick to the standard シ/ジ rather than write セィ/ゼィ or スィ/ズィ.

    I don't think Japanese speakers usually distinguish [si] from [ɕi] or [zi] from [ʑi], so there isn't a lot of use in distinguishing a "straight" /s/ or /z/ from the allophone that usually occurs before /i/. If you did write it with the less usual katakana combinations, I don't think people would pronounce it the way you intend very often.

So, depending on how you think the s and i should be pronounced:


I would use one of those.

| improve this answer | |
  • What do you think about コバ シン? – Earthliŋ Jul 4 '14 at 1:28
  • @Earthliŋ I know people insert /i/ instead of /u/ sometimes, but I went with /u/ because it's more common. – snailplane Jul 4 '14 at 1:29
  • I couldn't have asked for a better answer... I just want to know how you learned all this detailed information o.o... Very thorough, thank you. – Tek Jul 4 '14 at 18:45

There is no "correct" way. With some names there has become a standard way, such as スミス for Smith. In these circumstances the transition from foreign pronunciation is pretty straight forward.

The "u" in スミ is dropped in most pronunciations of the word making it similar to Sm and there is no closer way to represent this sound.

mi => ミ is straight forward.

ス is the closest approximation of the ending "th" sound in Japanese.

But for other names that are uncommon (i.e. if there's multiple katakana renditions of the word most likely almost no one would notice) then it comes down to coming up with a valid approximation.

Take the name Brekke.

Br => ブレ this is natural because again the "u" sound is dropped and the "re" is close to レ and the same when displayed using romaji.

The hard part is how to end the word. If you follow the romaji and use ブレッケー, which an acquaintance of mine had done by a company in Japan. However, the pronunciation of the name is closer to ブレッキー.

In this case, which is the "correct" version? Neither and both.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.