[apple/icu.git] / icuSources / data / translit / my_my_FONIPA.txt

# ***************************************************************************
# *
# *  Copyright (C) 2004-2016, International Business Machines
# *  Corporation; Unicode, Inc.; and others.  All Rights Reserved.
# *
# ***************************************************************************
# File: my_my_FONIPA.txt
# Generated from CLDR 
#

# Pronunciation rules for Burmese.
#
# The following rules are lexical and heuristic: lexical in the sense
# that they generate phoneme strings which may further undergo
# post-lexical phonological processes, in particular voicing, to
# result in actual surface forms; heuristic in the sense that they try
# to resolve ambiguities, especially around reduced vowels, in a
# systematic way that may be incorrect in many situations. Vowel
# reduction depends on many factors, such as morphemic structure,
# which are not available here.
#
# Definitions
#
# Dependent vowel signs
$vs_AA = \u102B;
$vs_aa = \u102C;
$vs_i = \u102D;
$vs_ii = \u102E;
$vs_u = \u102F;
$vs_uu = \u1030;
$vs_e = \u1031;
$vs_ai = \u1032;
# Various signs
$anusvara = \u1036;
$visarga = \u1038;
$virama = \u1039;
$asat = \u103A;
# Dependent (medial) consonant signs
$med_y = \u103B;
$med_r = \u103C;
$med_w = \u103D;
$med_h = \u103E;
# Independent letters and letter-like punctuation symbols
$independent = [\u1000-\u102A \u103F \u104C-\u104F \u1050-\u1055];
$creaky = \u0330;
$high = \u0301;
$low = \u0300;
$coda = [$creaky $high $low ɴ ʔ ə];  # TODO: remove if unused
#
# Preprocessing
#
::NFC;
# Replace U+102B TALL AA with U+102C AA. Their pronunciation is identical.
$vs_AA → $vs_aa;
# Unstack kinzi (င\u103A plus U+1039 VIRAMA) into plain င\u103A.
# Hmm, what would happen if the syllable ending in kinzi had non-low tone?
င\u103A $virama → င\u103A;
# Unstack everything else, i.e. replace U+1039 VIRAMA with U+103A ASAT.
$virama → $asat;
# Unstack U+103F GREAT SA.
ဿ → သ\u103Aသ;
# Insert a syllable boundary marker /./ before every independent letter.
::Null;
[^.$] { } $independent ([\u1037\u103B-\u103E])* [^\u103A] → \.;
# Insert default inherent vowel: /a\u0330/ at the end, /ə/ everywhere else.
::Null;
([\u1000-\u1021\u103F] [\u103B-\u103E]*) } [$] → $1 a $creaky;
([\u1000-\u1021\u103F] [\u103B-\u103E]*) } \.  → $1 ə;
# Allow for additional coda consonants.
#
# This only covers a few of the cases in which full coda consonants
# can appear in loanwords. The general situation is somewhat rare and
# is more easily dealt with in a formalism that can impose structural
# constraints on syllables more easily.
::Null;
$asat ($visarga)? [\u1000-\u102A] { $asat → ;
# Deal with ၎င\u103Aး early.
၎င\u103Aး → lə\.ɡa $high ʊ\u032Fɴ;
#
# Rhymes
#
::Null;
က\u103A → ɛʔ;
ဂ\u103A → ɛʔ;  # in မဂ\u1039ဂဇင\u103Aး ~ မဂ\u103Aဂဇင\u103Aး /mɛʔ.ɡə.zɪ\u0301ɴ/
င\u1037\u103A → ɪ $creaky ɴ;
င\u103Aး → ɪ $high ɴ;
င\u103A → ɪ $low ɴ;
စ\u103A → ɪʔ;  # maybe sometimes /eɪ\u032Fʔ/
ဉ\u1037\u103A → ɪ $creaky ɴ;
ဉ\u103Aး → ɪ $high ɴ;
ဉ\u103A → ɪ $low ɴ;
ည\u1037\u103A → ɛ $creaky;
ည\u103Aး → ɛ $high;
ည\u103A → ɛ $low;
ဏ\u1037\u103A → a $creaky ɴ;
ဏ\u103Aး → a $high ɴ;
ဏ\u103A → a $low ɴ;
တ\u103A → aʔ;
န\u1037\u103A → a $creaky ɴ;
န\u103Aး → a $high ɴ;
န\u103A → a $low ɴ;
ပ\u103A → aʔ;
မ\u1037\u103A → a $creaky ɴ;
မ\u103Aး → a $high ɴ;
မ\u103A → a $low ɴ;
ယ\u1037\u103A → ɛ $creaky;
ယ\u103Aး → ɛ $high;
ယ\u103A → ɛ $low;
သ\u103A → aʔ;
$vs_aa ဉ\u1037\u103A → ɪ $creaky ɴ;
$vs_aa ဉ\u103Aး → ɪ $high ɴ;
$vs_aa ဉ\u103A → ɪ $low ɴ;
$vs_aa တ\u103A → aʔ;
$vs_aa ဏ\u1037\u103A → a $creaky ɴ;
$vs_aa ဏ\u103Aး → a $high ɴ;
$vs_aa ဏ\u103A → a $low ɴ;
$vs_aa န\u1037\u103A → a $creaky ɴ;
$vs_aa န\u103Aး → a $high ɴ;
$vs_aa န\u103A → a $low ɴ;
$vs_aa ပ\u103A → aʔ;  # in ကလာပ\u103Aစည\u103Aး /kə.laʔ.sɛ\u0301/ (club cell)
$vs_aa ယ\u1037\u103A → ɛ $creaky;
$vs_aa ယ\u103Aး → ɛ $high;
$vs_aa ယ\u103A → ɛ $low;
$vs_aa \u1037 → a $creaky;  # redundant creaky tone
$vs_aa း → a $high;
$vs_aa → a $low;
$vs_i က\u103A → eɪ\u032Fʔ;
$vs_i စ\u103A → eɪ\u032Fʔ;
$vs_i တ\u103A → eɪ\u032Fʔ;
$vs_i န\u1037\u103A → e $creaky ɪ\u032Fɴ;
$vs_i န\u103Aး → e $high ɪ\u032Fɴ;
$vs_i န\u103A → e $low ɪ\u032Fɴ;
$vs_i ပ\u103A → eɪ\u032Fʔ;
$vs_i မ\u1037\u103A → e $creaky ɪ\u032Fɴ;
$vs_i မ\u103Aး → e $high ɪ\u032Fɴ;
$vs_i မ\u103A → e $low ɪ\u032Fɴ;
$vs_i $vs_u က\u103A → aɪ\u032Fʔ;
$vs_i $vs_u င\u1037\u103A → a $creaky ɪ\u032Fɴ;
$vs_i $vs_u င\u103Aး → a $high ɪ\u032Fɴ;
$vs_i $vs_u င\u103A → a $low ɪ\u032Fɴ;
$vs_i $vs_u ဏ\u1037\u103A → a $creaky ɪ\u032Fɴ;
$vs_i $vs_u ဏ\u103Aး → a $high ɪ\u032Fɴ;
$vs_i $vs_u ဏ\u103A → a $low ɪ\u032Fɴ;
$vs_i $vs_u ယ\u1037\u103A → o $creaky;
$vs_i $vs_u ယ\u103Aး → o $high;
$vs_i $vs_u ယ\u103A → o $low;  # in က\u102D\u102Fယ\u103A /kò/
$vs_i $vs_u \u1037 → o $creaky;
$vs_i $vs_u း → o $high;
$vs_i $vs_u → o $low;
$vs_i $anusvara \u1037 → e $creaky ɪ\u032Fɴ;
$vs_i $anusvara း → e $high ɪ\u032Fɴ;
$vs_i $anusvara → e $low ɪ\u032Fɴ;
$vs_i → i $creaky;
$vs_ii \u1037 → i $creaky;  # this does not usually occur
$vs_ii း → i $high;
$vs_ii → i $low;
$vs_u က\u103A → oʊ\u032Fʔ;
$vs_u ဂ\u103A → oʊ\u032Fʔ;
$vs_u ဏ\u1037\u103A → o $creaky ʊ\u032Fɴ;
$vs_u ဏ\u103Aး → o $high ʊ\u032Fɴ;
$vs_u ဏ\u103A → o $low ʊ\u032Fɴ;
$vs_u တ\u103A → oʊ\u032Fʔ;
$vs_u န\u1037\u103A → o $creaky ʊ\u032Fɴ;
$vs_u န\u103Aး → o $high ʊ\u032Fɴ;
$vs_u န\u103A → o $low ʊ\u032Fɴ;
$vs_u ပ\u103A → oʊ\u032Fʔ;
$vs_u မ\u1037\u103A → o $creaky ʊ\u032Fɴ;
$vs_u မ\u103Aး → o $high ʊ\u032Fɴ;
$vs_u မ\u103A → o $low ʊ\u032Fɴ;
$vs_u $anusvara \u1037 → o $creaky ʊ\u032Fɴ;
$vs_u $anusvara း → o $high ʊ\u032Fɴ;
$vs_u $anusvara → o $low ʊ\u032Fɴ;
$vs_u → u $creaky;
$vs_uu \u1037 → u $creaky;  # this does not usually occur
$vs_uu း → u $high;
$vs_uu → u $low;
$vs_e တ\u103A → ɪʔ;
$vs_e $vs_aa က\u103A → aʊ\u032Fʔ;
$vs_e $vs_aa င\u1037\u103A → a $creaky ʊ\u032Fɴ;
$vs_e $vs_aa င\u103Aး → a $high ʊ\u032Fɴ;
$vs_e $vs_aa င\u103A → a $low ʊ\u032Fɴ;
$vs_e $vs_aa \u1037 → ɔ $creaky;
$vs_e $vs_aa း → ɔ $high;  # redundant high tone; this does not usually occur
$vs_e $vs_aa \u103A → ɔ $low;
$vs_e $vs_aa → ɔ $high;
$vs_e \u1037 → e $creaky;
$vs_e း → e $high;
$vs_e → e $low;
$vs_ai \u1037 → ɛ $creaky;
$vs_ai း → ɛ $high;  # redundant high tone; this does not usually occur
$vs_ai → ɛ $high;
$anusvara \u1037 → a $creaky ɴ;
$anusvara း → a $high ɴ;
$anusvara → a $low ɴ;
$med_w တ\u103A → ʊʔ;
$med_w န\u1037\u103A → ʊ $creaky ɴ;
$med_w န\u103Aး → ʊ $high ɴ;
$med_w န\u103A → ʊ $low ɴ;
$med_w ပ\u103A → ʊʔ;
$med_w မ\u1037\u103A → ʊ $creaky ɴ;
$med_w မ\u103Aး → ʊ $high ɴ;
$med_w မ\u103A → ʊ $low ɴ;
#
# Medials
#
::Null;
# Palatalization of the velar stops before MEDIAL YA and MEDIAL RA:
# velar + /j/ ==> modern palatals.
ကျ → t\u0361ɕ;
ချ → t\u0361ɕʰ;
ဂျ → d\u0361ʑ;
ဃျ → d\u0361ʑ;
ကြ → t\u0361ɕ;
ခြ → t\u0361ɕʰ;
ဂြ → d\u0361ʑ;
ဃြ → d\u0361ʑ;
# Remove redundant MEDIAL YA and MEDIAL RA after initial YA.
ယ { [$med_y $med_r] → ;
# Reorder the medials so that U+103E SIGN MEDIAL HA comes before any
# other medials.
# First, push U+103E MEDIAL HA before U+103D MEDIAL WA.
\u103D \u103E → \u103E \u103D;
::Null;
# Now MEDIAL WA comes last.
# Produce the palatal ʃ from (SA|LA)+YA+HA.
သျ\u103E → ʃ;
လျ\u103E → ʃ;
# Second, push U+103E MEDIAL HA before U+103C MEDIAL RA.
\u103C \u103E → \u103E \u103C;
::Null;
# Finally, push U+103E MEDIAL HA before U+103B MEDIAL YA.
\u103B \u103E → \u103E \u103B;
::Null;
# Consume MEDIAL HA and apply devoicing.
င\u103E → ŋ\u030A;
ဉ\u103E → ɲ\u0325;
ည\u103E → ɲ\u0325;
ဏ\u103E → n\u0325;
န\u103E → n\u0325;
မ\u103E → m\u0325;
ယ\u103E → ʃ;
ရ\u103E → ʃ;
လ\u103E → l\u0325;
ဝ\u103E → w\u0325;
ဠ\u103E → l\u0325;
# Drop any remaining U+103E MEDIAL HA.
\u103E → ;
# Simplify medial cluster /jw/ to /w/, i.e. drop U+103B MEDIAL YA and
# U+103C MEDIAL RA before U+103D MEDIAL WA.  # TODO: revisit this
\u103B } \u103D → ;
\u103C } \u103D → ;
\u103B → j;
\u103C → j;
\u103D → w;
#
# Initials
#
# Velars
က → k;
ခ → kʰ;
ဂ → ɡ;
ဃ → ɡ;
င → ŋ;
# Historic palatals
စ → s;
ဆ → sʰ;
ဇ → z;
ဈ → z;
ဉ → ɲ;
ည → ɲ;
# Alveolars
ဋ → t;
ဌ → tʰ;
ဍ → d;
ဎ → d;
ဏ → n;
# Historic dentals ==> alveolars
တ → t;
ထ → tʰ;
ဒ → d;
ဓ → d;
န → n;
# Labials
ပ → p;
ဖ → pʰ;
ဗ → b;
ဘ → b;
မ → m;
# Other letters
ယ → j;
ရ → j;  # historic /r/
လ\u103A → ;  # final, typically not pronounced in native words
လ → l;
ဝ → w;
သ → θ;  # historic /s/ ==> modern dental
ဟ → h;
ဠ → l;
အ → ʔ;
# Independent vowels
ဣ\u1037 → ʔḭ;  # redundant creaky tone; this does not usually occur
ဣး → ʔí;  # this does not usually occur
ဣ → ʔḭ;
ဤ\u1037 → ʔḭ;  # this does not usually occur
ဤး → ʔí;  # this does not usually occur
ဤ → ʔì;
ဥ\u1037 → ʔṵ;  # redundant creaky tone; this does not usually occur
ဥး → ʔú;  # this does not usually occur
ဥ → ʔṵ;
ဦ\u1037 → ʔṵ;  # this does not usually occur
ဦး → ʔú;
ဦ → ʔù;
ဧ\u1037 → ʔḛ;  # this does not usually occur
ဧး → ʔé;
ဧ → ʔè;
ဩ\u1037 → ʔɔ\u0330;  # this does not usually occur
ဩး → ʔɔ\u0301;  # redundant high tone; this does not usually occur
ဩ → ʔɔ\u0301;
ဪ\u1037 → ʔɔ\u0330;  # this does not usually occur
ဪး → ʔɔ\u0301;  # this does not usually occur
ဪ → ʔɔ\u0300;
# Various signs
၌ → n\u0325aɪ\u032Fʔ;
၍ → jwḛ;
# ၎င\u103Aး was handled earlier.
၏ → ʔḭ;
#
# Postprocessing
#
# Delete any remaining U+103A ASAT.
$asat → ;
# Delete zero-width space, non-joiner, joiner.
[\u200B-\u200D] → ;
::NFC;
Commit	Line	Data
2ca993e8 A	1	# ***************************************************************************
	2	# *
	3	# * Copyright (C) 2004-2016, International Business Machines
	4	# * Corporation; Unicode, Inc.; and others. All Rights Reserved.
	5	# *
	6	# ***************************************************************************
	7	# File: my_my_FONIPA.txt
	8	# Generated from CLDR
	9	#
	10
	11	# Pronunciation rules for Burmese.
	12	#
	13	# The following rules are lexical and heuristic: lexical in the sense
	14	# that they generate phoneme strings which may further undergo
	15	# post-lexical phonological processes, in particular voicing, to
	16	# result in actual surface forms; heuristic in the sense that they try
	17	# to resolve ambiguities, especially around reduced vowels, in a
	18	# systematic way that may be incorrect in many situations. Vowel
	19	# reduction depends on many factors, such as morphemic structure,
	20	# which are not available here.
	21	#
	22	# Definitions
	23	#
	24	# Dependent vowel signs
	25	$vs_AA = \u102B;
	26	$vs_aa = \u102C;
	27	$vs_i = \u102D;
	28	$vs_ii = \u102E;
	29	$vs_u = \u102F;
	30	$vs_uu = \u1030;
	31	$vs_e = \u1031;
	32	$vs_ai = \u1032;
	33	# Various signs
	34	$anusvara = \u1036;
	35	$visarga = \u1038;
	36	$virama = \u1039;
	37	$asat = \u103A;
	38	# Dependent (medial) consonant signs
	39	$med_y = \u103B;
	40	$med_r = \u103C;
	41	$med_w = \u103D;
	42	$med_h = \u103E;
	43	# Independent letters and letter-like punctuation symbols
	44	$independent = [\u1000-\u102A \u103F \u104C-\u104F \u1050-\u1055];
	45	$creaky = \u0330;
	46	$high = \u0301;
	47	$low = \u0300;
	48	$coda = [$creaky $high $low ɴ ʔ ə]; # TODO: remove if unused
	49	#
	50	# Preprocessing
	51	#
	52	::NFC;
	53	# Replace U+102B TALL AA with U+102C AA. Their pronunciation is identical.
	54	$vs_AA → $vs_aa;
	55	# Unstack kinzi (င\u103A plus U+1039 VIRAMA) into plain င\u103A.
	56	# Hmm, what would happen if the syllable ending in kinzi had non-low tone?
	57	င\u103A $virama → င\u103A;
	58	# Unstack everything else, i.e. replace U+1039 VIRAMA with U+103A ASAT.
	59	$virama → $asat;
	60	# Unstack U+103F GREAT SA.
	61	ဿ → သ\u103Aသ;
	62	# Insert a syllable boundary marker /./ before every independent letter.
	63	::Null;
	64	[^.$] { } $independent ([\u1037\u103B-\u103E])* [^\u103A] → \.;
65	# Insert default inherent vowel: /a\u0330/ at the end, /ə/ everywhere else.
66	::Null;
67	([\u1000-\u1021\u103F] [\u103B-\u103E]*) } [$] → $1 a $creaky;
68	([\u1000-\u1021\u103F] [\u103B-\u103E]*) } \. → $1 ə;
69	# Allow for additional coda consonants.
70	#
71	# This only covers a few of the cases in which full coda consonants
72	# can appear in loanwords. The general situation is somewhat rare and
73	# is more easily dealt with in a formalism that can impose structural
74	# constraints on syllables more easily.
75	::Null;
76	$asat ($visarga)? [\u1000-\u102A] { $asat → ;
77	# Deal with ၎င\u103Aး early.
78	၎င\u103Aး → lə\.ɡa $high ʊ\u032Fɴ;
79	#
80	# Rhymes
81	#
82	::Null;
83	က\u103A → ɛʔ;
84	ဂ\u103A → ɛʔ; # in မဂ\u1039ဂဇင\u103Aး ~ မဂ\u103Aဂဇင\u103Aး /mɛʔ.ɡə.zɪ\u0301ɴ/
85	င\u1037\u103A → ɪ $creaky ɴ;
86	င\u103Aး → ɪ $high ɴ;
87	င\u103A → ɪ $low ɴ;
88	စ\u103A → ɪʔ; # maybe sometimes /eɪ\u032Fʔ/
89	ဉ\u1037\u103A → ɪ $creaky ɴ;
90	ဉ\u103Aး → ɪ $high ɴ;
91	ဉ\u103A → ɪ $low ɴ;
92	ည\u1037\u103A → ɛ $creaky;
93	ည\u103Aး → ɛ $high;
94	ည\u103A → ɛ $low;
95	ဏ\u1037\u103A → a $creaky ɴ;
96	ဏ\u103Aး → a $high ɴ;
97	ဏ\u103A → a $low ɴ;
98	တ\u103A → aʔ;
99	န\u1037\u103A → a $creaky ɴ;
100	န\u103Aး → a $high ɴ;
101	န\u103A → a $low ɴ;
102	ပ\u103A → aʔ;
103	မ\u1037\u103A → a $creaky ɴ;
104	မ\u103Aး → a $high ɴ;
105	မ\u103A → a $low ɴ;
106	ယ\u1037\u103A → ɛ $creaky;
107	ယ\u103Aး → ɛ $high;
108	ယ\u103A → ɛ $low;
109	သ\u103A → aʔ;
110	$vs_aa ဉ\u1037\u103A → ɪ $creaky ɴ;
111	$vs_aa ဉ\u103Aး → ɪ $high ɴ;
112	$vs_aa ဉ\u103A → ɪ $low ɴ;
113	$vs_aa တ\u103A → aʔ;
114	$vs_aa ဏ\u1037\u103A → a $creaky ɴ;
115	$vs_aa ဏ\u103Aး → a $high ɴ;
116	$vs_aa ဏ\u103A → a $low ɴ;
117	$vs_aa န\u1037\u103A → a $creaky ɴ;
118	$vs_aa န\u103Aး → a $high ɴ;
119	$vs_aa န\u103A → a $low ɴ;
120	$vs_aa ပ\u103A → aʔ; # in ကလာပ\u103Aစည\u103Aး /kə.laʔ.sɛ\u0301/ (club cell)
121	$vs_aa ယ\u1037\u103A → ɛ $creaky;
122	$vs_aa ယ\u103Aး → ɛ $high;
123	$vs_aa ယ\u103A → ɛ $low;
124	$vs_aa \u1037 → a $creaky; # redundant creaky tone
125	$vs_aa း → a $high;
126	$vs_aa → a $low;
127	$vs_i က\u103A → eɪ\u032Fʔ;
128	$vs_i စ\u103A → eɪ\u032Fʔ;
129	$vs_i တ\u103A → eɪ\u032Fʔ;
130	$vs_i န\u1037\u103A → e $creaky ɪ\u032Fɴ;
131	$vs_i န\u103Aး → e $high ɪ\u032Fɴ;
132	$vs_i န\u103A → e $low ɪ\u032Fɴ;
133	$vs_i ပ\u103A → eɪ\u032Fʔ;
134	$vs_i မ\u1037\u103A → e $creaky ɪ\u032Fɴ;
135	$vs_i မ\u103Aး → e $high ɪ\u032Fɴ;
136	$vs_i မ\u103A → e $low ɪ\u032Fɴ;
137	$vs_i $vs_u က\u103A → aɪ\u032Fʔ;
138	$vs_i $vs_u င\u1037\u103A → a $creaky ɪ\u032Fɴ;
139	$vs_i $vs_u င\u103Aး → a $high ɪ\u032Fɴ;
140	$vs_i $vs_u င\u103A → a $low ɪ\u032Fɴ;
141	$vs_i $vs_u ဏ\u1037\u103A → a $creaky ɪ\u032Fɴ;
142	$vs_i $vs_u ဏ\u103Aး → a $high ɪ\u032Fɴ;
143	$vs_i $vs_u ဏ\u103A → a $low ɪ\u032Fɴ;
144	$vs_i $vs_u ယ\u1037\u103A → o $creaky;
145	$vs_i $vs_u ယ\u103Aး → o $high;
146	$vs_i $vs_u ယ\u103A → o $low; # in က\u102D\u102Fယ\u103A /kò/
147	$vs_i $vs_u \u1037 → o $creaky;
148	$vs_i $vs_u း → o $high;
149	$vs_i $vs_u → o $low;
150	$vs_i $anusvara \u1037 → e $creaky ɪ\u032Fɴ;
151	$vs_i $anusvara း → e $high ɪ\u032Fɴ;
152	$vs_i $anusvara → e $low ɪ\u032Fɴ;
153	$vs_i → i $creaky;
154	$vs_ii \u1037 → i $creaky; # this does not usually occur
155	$vs_ii း → i $high;
156	$vs_ii → i $low;
157	$vs_u က\u103A → oʊ\u032Fʔ;
158	$vs_u ဂ\u103A → oʊ\u032Fʔ;
159	$vs_u ဏ\u1037\u103A → o $creaky ʊ\u032Fɴ;
160	$vs_u ဏ\u103Aး → o $high ʊ\u032Fɴ;
161	$vs_u ဏ\u103A → o $low ʊ\u032Fɴ;
162	$vs_u တ\u103A → oʊ\u032Fʔ;
163	$vs_u န\u1037\u103A → o $creaky ʊ\u032Fɴ;
164	$vs_u န\u103Aး → o $high ʊ\u032Fɴ;
165	$vs_u န\u103A → o $low ʊ\u032Fɴ;
166	$vs_u ပ\u103A → oʊ\u032Fʔ;
167	$vs_u မ\u1037\u103A → o $creaky ʊ\u032Fɴ;
168	$vs_u မ\u103Aး → o $high ʊ\u032Fɴ;
169	$vs_u မ\u103A → o $low ʊ\u032Fɴ;
170	$vs_u $anusvara \u1037 → o $creaky ʊ\u032Fɴ;
171	$vs_u $anusvara း → o $high ʊ\u032Fɴ;
172	$vs_u $anusvara → o $low ʊ\u032Fɴ;
173	$vs_u → u $creaky;
174	$vs_uu \u1037 → u $creaky; # this does not usually occur
175	$vs_uu း → u $high;
176	$vs_uu → u $low;
177	$vs_e တ\u103A → ɪʔ;
178	$vs_e $vs_aa က\u103A → aʊ\u032Fʔ;
179	$vs_e $vs_aa င\u1037\u103A → a $creaky ʊ\u032Fɴ;
180	$vs_e $vs_aa င\u103Aး → a $high ʊ\u032Fɴ;
181	$vs_e $vs_aa င\u103A → a $low ʊ\u032Fɴ;
182	$vs_e $vs_aa \u1037 → ɔ $creaky;
183	$vs_e $vs_aa း → ɔ $high; # redundant high tone; this does not usually occur
184	$vs_e $vs_aa \u103A → ɔ $low;
185	$vs_e $vs_aa → ɔ $high;
186	$vs_e \u1037 → e $creaky;
187	$vs_e း → e $high;
188	$vs_e → e $low;
189	$vs_ai \u1037 → ɛ $creaky;
190	$vs_ai း → ɛ $high; # redundant high tone; this does not usually occur
191	$vs_ai → ɛ $high;
192	$anusvara \u1037 → a $creaky ɴ;
193	$anusvara း → a $high ɴ;
194	$anusvara → a $low ɴ;
195	$med_w တ\u103A → ʊʔ;
196	$med_w န\u1037\u103A → ʊ $creaky ɴ;
197	$med_w န\u103Aး → ʊ $high ɴ;
198	$med_w န\u103A → ʊ $low ɴ;
199	$med_w ပ\u103A → ʊʔ;
200	$med_w မ\u1037\u103A → ʊ $creaky ɴ;
201	$med_w မ\u103Aး → ʊ $high ɴ;
202	$med_w မ\u103A → ʊ $low ɴ;
203	#
204	# Medials
205	#
206	::Null;
207	# Palatalization of the velar stops before MEDIAL YA and MEDIAL RA:
208	# velar + /j/ ==> modern palatals.
209	ကျ → t\u0361ɕ;
210	ချ → t\u0361ɕʰ;
211	ဂျ → d\u0361ʑ;
212	ဃျ → d\u0361ʑ;
213	ကြ → t\u0361ɕ;
214	ခြ → t\u0361ɕʰ;
215	ဂြ → d\u0361ʑ;
216	ဃြ → d\u0361ʑ;
217	# Remove redundant MEDIAL YA and MEDIAL RA after initial YA.
218	ယ { [$med_y $med_r] → ;
219	# Reorder the medials so that U+103E SIGN MEDIAL HA comes before any
220	# other medials.
221	# First, push U+103E MEDIAL HA before U+103D MEDIAL WA.
222	\u103D \u103E → \u103E \u103D;
223	::Null;
224	# Now MEDIAL WA comes last.
225	# Produce the palatal ʃ from (SA\|LA)+YA+HA.
226	သျ\u103E → ʃ;
227	လျ\u103E → ʃ;
228	# Second, push U+103E MEDIAL HA before U+103C MEDIAL RA.
229	\u103C \u103E → \u103E \u103C;
230	::Null;
231	# Finally, push U+103E MEDIAL HA before U+103B MEDIAL YA.
232	\u103B \u103E → \u103E \u103B;
233	::Null;
234	# Consume MEDIAL HA and apply devoicing.
235	င\u103E → ŋ\u030A;
236	ဉ\u103E → ɲ\u0325;
237	ည\u103E → ɲ\u0325;
238	ဏ\u103E → n\u0325;
239	န\u103E → n\u0325;
240	မ\u103E → m\u0325;
241	ယ\u103E → ʃ;
242	ရ\u103E → ʃ;
243	လ\u103E → l\u0325;
244	ဝ\u103E → w\u0325;
245	ဠ\u103E → l\u0325;
246	# Drop any remaining U+103E MEDIAL HA.
247	\u103E → ;
248	# Simplify medial cluster /jw/ to /w/, i.e. drop U+103B MEDIAL YA and
249	# U+103C MEDIAL RA before U+103D MEDIAL WA. # TODO: revisit this
250	\u103B } \u103D → ;
251	\u103C } \u103D → ;
252	\u103B → j;
253	\u103C → j;
254	\u103D → w;
255	#
256	# Initials
257	#
258	# Velars
259	က → k;
260	ခ → kʰ;
261	ဂ → ɡ;
262	ဃ → ɡ;
263	င → ŋ;
264	# Historic palatals
265	စ → s;
266	ဆ → sʰ;
267	ဇ → z;
268	ဈ → z;
269	ဉ → ɲ;
270	ည → ɲ;
271	# Alveolars
272	ဋ → t;
273	ဌ → tʰ;
274	ဍ → d;
275	ဎ → d;
276	ဏ → n;
277	# Historic dentals ==> alveolars
278	တ → t;
279	ထ → tʰ;
280	ဒ → d;
281	ဓ → d;
282	န → n;
283	# Labials
284	ပ → p;
285	ဖ → pʰ;
286	ဗ → b;
287	ဘ → b;
288	မ → m;
289	# Other letters
290	ယ → j;
291	ရ → j; # historic /r/
292	လ\u103A → ; # final, typically not pronounced in native words
293	လ → l;
294	ဝ → w;
295	သ → θ; # historic /s/ ==> modern dental
296	ဟ → h;
297	ဠ → l;
298	အ → ʔ;
299	# Independent vowels
300	ဣ\u1037 → ʔḭ; # redundant creaky tone; this does not usually occur
301	ဣး → ʔí; # this does not usually occur
302	ဣ → ʔḭ;
303	ဤ\u1037 → ʔḭ; # this does not usually occur
304	ဤး → ʔí; # this does not usually occur
305	ဤ → ʔì;
306	ဥ\u1037 → ʔṵ; # redundant creaky tone; this does not usually occur
307	ဥး → ʔú; # this does not usually occur
308	ဥ → ʔṵ;
309	ဦ\u1037 → ʔṵ; # this does not usually occur
310	ဦး → ʔú;
311	ဦ → ʔù;
312	ဧ\u1037 → ʔḛ; # this does not usually occur
313	ဧး → ʔé;
314	ဧ → ʔè;
315	ဩ\u1037 → ʔɔ\u0330; # this does not usually occur
316	ဩး → ʔɔ\u0301; # redundant high tone; this does not usually occur
317	ဩ → ʔɔ\u0301;
318	ဪ\u1037 → ʔɔ\u0330; # this does not usually occur
319	ဪး → ʔɔ\u0301; # this does not usually occur
320	ဪ → ʔɔ\u0300;
321	# Various signs
322	၌ → n\u0325aɪ\u032Fʔ;
323	၍ → jwḛ;
324	# ၎င\u103Aး was handled earlier.
325	၏ → ʔḭ;
326	#
327	# Postprocessing
328	#
329	# Delete any remaining U+103A ASAT.
330	$asat → ;
331	# Delete zero-width space, non-joiner, joiner.
332	[\u200B-\u200D] → ;
333	::NFC;
334