Golden Rule of becoming a native Cantonese speaker: Tones > Everything!

In this blog, we use Jyutping to indicate the pronunciation of Cantonese characters.

Previous lesson: Rimes with a & aa

All Rimes with e

In this lesson, we talk about Rimes with e. e is a very naughty sound because it has multiple sounds based on how you spell it.

Also, e can combine with o to create oe and eo sounds.

Rimes with e

Jyutping	Sounds in English	Cantonese Example	Jyutping	Sounds in English	Cantonese Example
e	ai in pair	些(se1)	eng	eng in leng2	鄭(zeng6)
ei	ey in prey	四(sei3)	ep	ep in bicep	夾(gaap3)
eu	eil in veil	掉(diu6)	et	et in pet	坺(paat6)
em	em in gem	舔(tim2)	ek	ek in trek	石(sek6)

We can see most of them are the same as e pronounced in English.

Rimes with oe and eo

Golden Rule of becoming a native Cantonese speaker: Tones > Everything!

In this blog, we use Jyutping to indicate the pronunciation of Cantonese characters.

Previous lesson: Introduction

Recap of Rimes

Rimes is a syllable of vowels with consonants that you can rhyme!

For example, in English, the Rime of “Cow” is “ow”; in Cantonese, cow is 牛(ngau4), and the Rime is “au”.

Difference of a and aa

Two vowels in Cantonese start with a. a sounds like “uh” in English, and aa “ahhh” in English.

And all their combinations as well. For example, 分(fan1) sounds like “fun”, and 飯(faan6) sounds like “fran” in France without “r” sound.

Still, it may be difficult to distinguish their difference when they form rimes with other sounds.

Golden Rule of becoming a native Cantonese speaker: Tones > Everything!

In this blog, we use Jyutping to indicate the pronunciation of Cantonese characters.

Previous lesson: Introduction

Introduction of Onset

In this lesson, We start with Onset. They are all consonants that are used before vowels.

For example, “good morning” in Cantonese is 早晨(zou2 san4), where “z” and “s” are Onset here.

Onset

Basically, these consonant looks identical, or you may find alternative spelling in English, except
kw, z, and c might be a little unique.

Jyutping	Sounds in English	Cantonese Example	Jyutping	Sounds in English	Cantonese Example
-	character that have no starting consonant (Null initial)	呀(aa3)	k	k in king	卡(kaa1)
b	b in bar	巴(baa1)	ng	ng in sing	牙(ngaa4)
p	p in palm	怕(paa3)	h	h in harp	蝦(haa1)
m	m in mat	媽(maa1)	gw	gu in guava	瓜(gwaa1)
f	f in foul	花(faa1)	kw	qu in aqua	誇(kwaa1)
d	d in dip1	打(daa2)	w	w in wow	蛙(waa1)
t	t in tip1	他(taa1)	z	j in job but with a ‘t’ sound in front of it2	揸(zaa1)
n	n in nap	那(naa5)	c	c in chat2 without ‘h’ sound	叉(caa1)
l	l in lap	啦(laa1)	s	s in soup	沙(saa1)
g	g in gum	家(gaa1)	j	y in yes3	也(jaa5)

¹If you are a native English speaker, you may notice that the Cantonese “d” sound is softer than in English, but the “t” sound is more challenging (the difference is negligible here). But if you have a problem pronouncing this pair, try to make your tongue touch the upper front teeth when creating the “d” and “t” sounds.

Golden Rule of becoming a native Cantonese speaker: Tones > Everything!

In this blog, we use Jyutping to indicate the pronunciation of Cantonese characters.

Overview of Cantonese Phonology

Maybe you already know that we use one syllable for each character in Cantonese,
just like all other Chinese languages. It contains 3 crucial parts in order to pronounce the word.
That is the consonances, vowels, and tones.

In Cantonese, there are 19 consonances, 9 vowels and 6 tones.
That creates 1,760 different sounds to cover over 10,000 Chinese Characters.

Every Cantonese consonance and vowel pair can always be described as an Onset-Rime.

Phonology Structure

So what is actually an Onset-Rimes?
An Onset-Rimes is either a consonant-vowel or a consonant-vowel-consonant structure, For example:

• consonant-vowel: the word “rich” = 富(fu3) (sounds like “foo” in English).

• consonant-vowel-consonant: the word “bamboo” = 竹(zuk1) (sounds like “joke” in English).

Tones

This is a learning note of a course in CISPA, UdS. Taught by Swen Jacobs

Last post: Labeled Transition Systems

Processes

A Process is an LTS $P = (Q, Q_0, \Sigma, \delta, \lambda)$ with $\Sigma = \Sigma_{int} \cup \lbrace out_a , in_a \mid a \in \Sigma_{sync} \rbrace $, where

• $\Sigma_{int}$ is a set of internal actions,
• $\Sigma_{sync}$ a set of synchronizing actions,
• $out_a$ is an send action (or initiate action),
• $in_a$ is an receive action.

Composition by Synchronization

The composition $P^n$ of $n$ (uniform) processes wrt. $card$ is the LTS $(S, S_0, \Sigma_{int} \cup \Sigma_{sync}, \Delta, \Lambda)$ with:

• $S = Q \times Q$

• $S_0 = Q_0 \times Q_0$

• $\Delta \subseteq S \times (\Delta_{int} \cup \Delta_{sync} \times S)$ is the set of all transitions that satisfy one of the following:

  Internal Transition:

  • For some $i \in\lbrace 1, … , n \rbrace$ , $(s, a, s’) \in S \times\Sigma_{int}\times S$ is an element such that:
    $(s(i), a, s’(i)) \in\delta$ , and $s(j) = s’(j)$ for $i \ne j \in\lbrace 1, …, n \rbrace$

  Process $s(i)$ take the action, other processes $s(j)$ remain their current states.

  Synchronous Transition

  • For some $i \in \lbrace 1, … , n \rbrace$ and some $I \subseteq \lbrace 1, …, n \rbrace \setminus \lbrace i \rbrace$ with $|I| \in card$ ,
    $(s, a, s’) \in S \times \Sigma_{int} \times S$ is an element such that:

    • $s(i) \buildrel out_a \over\longrightarrow s’(i)$ is (a local transition) in $P$

      One process $s(i)$ take the send action.

    • for every $j \in I, s(j) \buildrel in_a \over\longrightarrow s’(j)$ is (a local transition) in $P$

      $I$ is the set of processes that can take the receive action. (size of $I$ must not be larger then $card$)

    • for every $j \in\lbrace 1, …, n \rbrace \setminus (I \cup\lbrace i \rbrace), s’(j) = s(j)$

      Other processes $s(j)$ that cannot take receive actions remain their current states.

    • $I$ is maximal.

      There does not exist a larger set $I’ \supset I$ with $|I’| \in card$ that for all $j \in I’$ ,
      there is a local transition from $s(j)$ that can take the receive action.

• $\Lambda(s) = \lbrace p_{i} \mid p \in \text{AP and p} \in\lambda(s(i)), i \in \lbrace 1, …, n\rbrace \rbrace$

In this course, send action = $out_a$ = $a!!$ ; receive action = $in_a$ = $a??$.

Example of Composition by Broadcast synchronization. $(card = \lbrace 1 \rbrace)$

This is a learning note of a course in CISPA, UdS. Taught by Swen Jacobs

Labeled Transition Systems (LTS)

LTS is a concept in theoretical computer science used in the study of computation. It is used to describe the potential behavior of discrete systems, e.g. Model Checking.

It consists of states and transitions between states, which may be labeled with labels chosen from a set; the same label may appear on more than one transition. Mathematically, it can be described as directed graph.

Labels

Labels can be used to describe the behaviour or distinguish between states.
If the set of label is a singleton (only contains one label), then it can be omitted in the system.

LTS v.s. Finite-state Automata

In Transition systems:

• The set of states is not necessarily finite, or even countable.
• The set of transitions is not necessarily finite, or even countable.
• No “start” state or “final” states are given.

Formal definition