There are a lot of words that if you were to sound out each individual character you will get wrong. For instance, student is がくせい . Is the only way to know how to say words to listen to actual natives? If I were to pronounce this it would sound something like, “gakusee” as I would sound out, “ku”. However, it’s actually more like, “gaksee”. Is there any rules to know how to correctly pronounce stuff like that, or do you just learn with time?
5 comments
Yes. To learn to speak Japanese you gotta hear japanese
But eventually you’ll find a pattern intuitively.
You pick this up naturally as you listen to lots of Japanese.
If you need to hear a specific word, type it into Forvo: https://forvo.com
Yeah in Japanese, when certain short vowels come between two “unvoiced consonants” (consonants that you don’t activate your voice box to pronounce eg. s, t, k, etc), or at the end of a word after an unvoiced consonant, the vowels become unvoiced. The vowels are still there, and they are still pronounced, but they are unvoiced, so they sound like they have been dropped (unless you’re used to hearing unvoiced vowels). The reason it sounds like they’ve been dropped to an English speaker is because there are no unvoiced vowels in English.
This is devoicing – here’s a pretty good video on the topic: [https://youtu.be/iYQM7BhJJns?t=57](https://youtu.be/iYQM7BhJJns?t=57)
I agree with the other commenters that you should definitely listen to Japanese to get a sense for the sound and rules intuitively, but there is a rule for this particular sound change. Namely, the i and u vowels get devoiced or even deleted between unvoiced consonants or after an unvoiced consonant before a pause. Unvoiced consonants are k, t, p, s, h (basically the for the hiragana that come in pairs with ゛, the ones without the dots are unvoiced).
See the Wikipedia article on the phenomenon for more details https://en.m.wikipedia.org/wiki/Japanese_phonology#Devoicing
I should note that usually there are not two devoiced consonants in a row, pitch accent can affect devoicing (and vice versa), and apparently in some contexts o and a can also get devoiced, so this is a general rule which is very useful, but you still need to do a lot of listening to how people actually pronounce words to get a feel for this, because there are exceptions.