- 1 Introduction
- 2 Cantonese typing video
- 3 Typing Cantonese characters – the current state
- 4 Google Input for Cantonese
- 5 Jyutping conversion table
Today, I’m going to take a break from writing super long posts, and do a quick, but hopefully, helpful post on Cantonese typing, a subject that almost every learner brings up, and how to use Google’s Cantonese Input method to type Cantonese.
(The reason I’m writing this is because Cantonese has been added to the iOS keyboard, which is very exciting to this community! Please check out the video and the links below to see where to grab the keyboards on different platforms.)
Cantonese typing video
(If this video looks familiar, it should be! I’ve included it in the Complete Guide To Learn Cantonese, along with over 50 other videos.)
Remember you can get Google’s Cantonese Input software in all Google apps, in Chrome, in iOS AND in Android, so there’s absolutely no reason to not at least try it out! Here are the links to the relevant pages.
Google account settings: https://myaccount.google.com/inputtools
Android Google Cantonese Input: https://play.google.com/store/apps/details?id=com.google.android.apps.inputmethod.cantonese&hl=en
Google has since updated a few of its apps, so these screenshots will hopefully reflect some changes (until they’re displaced by future changes).
Gboard → Languages → Chinese (Hong Kong [China]) → Cantonese
Google input tools
Chrome extension settings
Typing Cantonese characters – the current state
For the longest period of time, Cantonese learners (and to a certain extent Cantonese natives) have struggled with typing Cantonese out, and not without good reason…typing Cantonese is really hard!
The most popular input methods in Hong Kong include: 九宮 (gau2 gung1)、倉頡 (cong1 kit3)、速成 (cuk1 sing4). Unfortunately, these methods are either difficult or impractical to use for learners.
Cantonese input method: 倉頡
倉頡 is a system of input based on the concept of splitting a character into divisible components. For example, the character 輸 (syu1) can be split into five components, which are highlighted in red, yellow, green, light blue and blue respectively.
They’re represented like so:
Light blue: 橫
交 (gaau1) means “a cross”, resembling the component that look like the character 十.
人 (jan4) means “person”, and could refer either to the radical or a component that looks like the radical.
橫 (waang4) refers to a horizontal line, as seen with the light blue line.
勾 (ngau1) represents a hook, which is probably similar to the stroke name.
These are, in turn, represented by these characters:
交 = 十
人 = 人
橫 = 一
勾 = 弓
And these characters are, in turn, represented by these keys
十 = J
人 = O
一 = M
弓 = N
Generally, these representative keys are separated into five categories, as shown colour coded in the photo:
哲理 (zit3 lei5): philosophical / meaning
筆畫 (bat1 waak6): stroke
人體 (jan4 tai2): human body
字形 (zi6 jing4): character shape
難字 (naan4 zi6): difficult characters (miscellaneous)
I won’t bore you with the details, but these are the different ways in which character components are represented.
In other words, as a learner, you need to remember these combinations for each character, or you must have a very good intuition of how the components shown on these keys relate to the shape of each character.
Cantonese input method: 速成
速成 (cuk1 sing4) is a slight improvement on the above, requiring that you only input the first and the last component keys.
But that’s still a lot of memorization!
Cantonese input method: 九宮
With 九宮 (gau2 gung1), you’ve got something that’s slightly better in terms of usability. Instead of relying on your memory, you “type” out a character similar to how you’d “write” a character. In other words, it’s based on strokes.
Unlike writing a character by hand, the basic principle of 九宮 is:
The first two strokes on the left, the first stroke on the right.
With just three strokes, you’re supposed to be able to type any Chinese character, based on nine squares that represent commonly used components in Chinese characters.
With the character 新, for example, you would type in the dash, horizontal line, and diagonal line, which correspond to 4, 7, 9 in the numpad. You’d then select the character among the nine that show up.
While this is a lot easier to learn, this method is a lot slower, because you constantly have to look to choose the character you want.
So not so good for practical uses either!
Google Input for Cantonese
The best way for inputting Chinese, I have to conclude, would be to type based on sound. After all, inputting this way:
- Has NO learning curve (you already know how each word is pronounced);
- Is a lot more accurate (because you can input multiple characters, making the combination more unique to search)
Yup, Google’s done it again with Google Input!
To be more precise, Google Input for Cantonese has existed for a long time, but I feel like there aren’t no tutorials showing you how to use it, and it doesn’t seem that popular (despite its ease of use).
So I decided to create one here.
Jyutping conversion table
To be precise, the reason there’s a learning curve is because the Google Input method for Cantonese isn’t the same as Jyutping. It’s remarkably similar in many ways, but it’s not exactly the same thing.
So what this tutorial will do is to give you plenty of examples helping you map out all the initials and finals in Jyutping to the standard used by Google Input.
|Initial||Sample word||Jyutping||Google Input|
|b||八百||baat3 baak3||baat baak|
|c||察覺||caak3 gok3||chaat gok|
|d||大人||daai6 jan4||daai yan|
|f||蜂蜜||fung1 mat6||fung mat|
|g||各自||gok3 zi6||gok ji|
|gw||果仁||gwo2 jan4||gwo yan|
|h||好運||hou2 wan6||hou wan|
|j||也許||jaa5 heoi2||ya heui|
|k||卡片||kaat1 pin2||kaat pin|
|kw||誇張||kwaa1 zoeng1||kwaa jeung|
|l||蘿蔔||lo4 baak3||lo baak|
|m||貓王||maau1 wong4||maau wong|
|n||納稅||naap6 seoi3||naap seui|
|ng||我國||ngo5 gwok3||ngo gwok|
|p||鋪頭||pou3 tau2||pou tau|
|s||稍為||saau2 wai4||saau wai|
|t||頭條||tau4 tiu4||tau tiu|
|w||委屈||wai2 wat1||wai wat|
|z||作曲||zok3 kuk1||jok kuk|
|Final||Sample word||Jyutping||Google Input|
|aa||花瓶||faa1 ping4||fa ping|
|aai||差人||caai1 jan4||chaai yan|
|aau||抄牌||caau1 paai4||chaau paai|
|aam||參觀||caam1 gun1||chaam gun|
|aan||餐廳||caan1 teng1||chaan teng|
|aang||橙汁||caang2 zap1||chaang jap|
|aap||插水||caap3 seoi2||chaap seui|
|aat||刮花||gwaat3 faa1||gwaat fa|
|aak||客人||haak3 jan4||haak yan|
|ai||鬼神||gwai2 san4||gwai san|
|au||狗尾||gau2 mei5||gau mei|
|am||柑橘||gam1 gat1||gam gat|
|an||跟住||gan1 zyu6||gan jyu|
|ang||藤條||tang4 tiu2||tang tiu|
|ap||濕透||sap1 tau3||sap tau|
|at||乜嘢||mat1 je5||mat ye|
|ak||麥片||mak6 pin3||mak pin|
|e||車頭||ce1 tau4||che tau|
|ei||飛機||fei1 gei1||fei gei|
|eu||掉嘢||deu6 je5||*diu ye|
|eng||靚車||leng3 ce1||leng che|
|ep||夾到||gep3 dou3||*gip dou|
|ek||錫錫||sek3 sek3||sek sek|
|i||癡線||ci1 sin3||ci sin|
|iu||料理||liu6 lei5||liu lei|
|im||劍身||gim3 san1||gim san|
|in||見到||gin3 dou2||gin dou|
|ing||敬佩||ging3 pui3||ging pui|
|ip||怯場||hip3 coeng4||hip cheung|
|it||跌倒||dit3 dou2||dit dou|
|ik||擊敗||gik1 baai6||gik baai|
|o||波子||bo1 zi2||bo ji|
|oi||改善||goi2 sin6||goi sin|
|ou||傲慢||ngou6 maan6||ngou maan|
|on||幹事||gon3 si6||gon si|
|ong||康樂||hong1 lok6||hong lok|
|ot||割讓||got3 joeng6||got yeung|
|ok||角度||gok3 dou6||gok dou|
|oe||鋸扒||goe3 paa2||geu pa|
|oeng||薑茶||goeng1 caa4||geung cha|
|oek||腳臭||goek3 cau2||geuk chau|
|eoi||居住||geoi1 zyu6||geui jyu|
|eon||輪流||leon4 lau4||leun lau|
|eot||律政||leot6 zing3||leut jing|
|u||扶搖||fu4 jiu4||fu yiu|
|ui||晦氣||fui3 hei3||fui hei|
|un||歡樂||fun1 lok6||fun lok|
|ung||豐富||fung1 fu3||fung fu|
|ut||闊度||fut3 dou6||fut dou|
|uk||福祉||fuk1 zi2||fuk ji|
|yu||署長||cyu5 zoeng2||chyu jeung|
|yun||存在||cyun4 zoi6||chyun joi|
|yut||奪得||dyut6 dak1||dyut dak|
- 掉嘢 (deu6) is a sound that doesn’t exist in the system Google Input uses. Fortunately, there are very few words with this sound, so you can just hand write this. The reason diu is put in here is because there are multiple readings of this character, and diu is one of them.
- 舐 (lem2) is also a word with very few corresponding characters, and is a modern sound. This probably needs to be handwritten too.
- 夾到 (gep6 dou2) – this is another case of not having the “ep” sound in this input system. This character can be handwritten or input with another sound, “gip”.
Note that Google’s Input system is actually pretty tolerant – most of the time, if you input Jyutping instead of this system, it’ll still be able to recognize the word.
However, there are a few that you absolutely can’t mix up, and those are:
- c → ch
- j → y
- z → j
- oe → eu
- oeng → eung
- oek → euk
- eoi → eui
- eon → eun
- eot → eut
Helpful tip: For last characters in a word, you can just type the first letter of the romanization. For example, you can type “gwong dung w” instead of “gwong dung wa” 廣東話. It doesn’t seem a lot in this example, but it adds up when you start typing Cantonese all the time.
Is it just me, or do the above seem like Jyutping to Yale conversions? (Sorry, I don’t use Yale a lot.)
But, that’s it!
If you remember this table (just the places where they’re different), then you can get started with Google’s Input method for Cantonese right away.