scriptconv 0.0.3a20__tar.gz → 0.0.3a22__tar.gz

{scriptconv-0.0.3a20 → scriptconv-0.0.3a22}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scriptconv
-Version: 0.0.3a20
+Version: 0.0.3a22
 Summary: Zero-dependency script & phoneme-notation core — ISO-15924 detection & metadata, IPA↔ARPABET/X-SAMPA/Lexique/Kirshenbaum/Cotovía/RFE, Buckwalter↔Arabic, Hangul→jamo, kana
 License: Apache-2.0
 Project-URL: Homepage, https://github.com/TigreGotico/scriptconv
@@ -102,6 +102,7 @@ Requires-Dist: deep-phonemizer; extra == "en-phonemizers"
 Provides-Extra: ar-phonemizers
 Requires-Dist: scriptconv[phonemizers]; extra == "ar-phonemizers"
 Requires-Dist: arbtok; extra == "ar-phonemizers"
+Requires-Dist: ovos-number-parser; extra == "ar-phonemizers"
 Provides-Extra: shami
 Requires-Dist: scriptconv[phonemizers]; extra == "shami"
 Requires-Dist: regex; extra == "shami"
@@ -113,6 +114,7 @@ Requires-Dist: pypinyin<1,>=0.50; extra == "test"
 Requires-Dist: quebra-frases; extra == "test"
 Requires-Dist: langcodes; extra == "test"
 Requires-Dist: regex; extra == "test"
+Requires-Dist: ovos-number-parser; extra == "test"
 Dynamic: license-file
 
 # scriptconv

{scriptconv-0.0.3a20 → scriptconv-0.0.3a22}/pyproject.toml

@@ -57,9 +57,9 @@ ja-phonemizers = ["scriptconv[phonemizers]", "pyopenjtalk", "cutlet", "pykakasi>
 ko = ["scriptconv[phonemizers]", "g2pk", "regex"]
 zh-phonemizers = ["scriptconv[phonemizers]", "jieba", "pypinyin>=0.50,<1", "xpinyin", "g2pM", "pinyin_to_ipa"]
 en-phonemizers = ["scriptconv[phonemizers]", "g2p_en", "openphonemizer", "deep-phonemizer"]
-ar-phonemizers = ["scriptconv[phonemizers]", "arbtok"]
+ar-phonemizers = ["scriptconv[phonemizers]", "arbtok", "ovos-number-parser"]
 shami = ["scriptconv[phonemizers]", "regex"]
-test = ["pytest", "pytest-timeout", "pykakasi>=2.3,<3", "pypinyin>=0.50,<1", "quebra-frases", "langcodes", "regex"]
+test = ["pytest", "pytest-timeout", "pykakasi>=2.3,<3", "pypinyin>=0.50,<1", "quebra-frases", "langcodes", "regex", "ovos-number-parser"]
 
 [project.urls]
 Homepage = "https://github.com/TigreGotico/scriptconv"
@@ -72,6 +72,8 @@ include = ["scriptconv*"]
 scriptconv = ["py.typed"]
 "scriptconv.data" = ["*.tsv.gz"]
 "scriptconv.phonemizers._thirdparty" = ["ko_tables/*.csv", "ko_tables/*.tsv"]
+"scriptconv.phonemizers._vendored.kog2p" = ["rulebook.txt", "LICENSE.md"]
+"scriptconv.phonemizers._vendored.mantoq" = ["LICENSE.md"]
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]

scriptconv-0.0.3a22/scriptconv/phonemizers/_vendored/__init__.py

@@ -0,0 +1,16 @@
+"""QUARANTINED vendored code — NOT covered by scriptconv's Apache-2.0 license.
+
+Each subpackage here is unpublished upstream code vendored for usability,
+distributed under ITS OWN license, stated in its directory:
+
+* ``mantoq/`` — Arabic G2P (github.com/mush42/mantoq). Its phonetisation core
+  (``buck/phonetise_buckwalter.py``) is **CC BY-NC 4.0** (Nawar Halabi's
+  Arabic-Phonetiser) — non-commercial use only. See ``mantoq/LICENSE.md``.
+* ``kog2p/`` — Korean G2P (github.com/scarletcho/KoG2P), **GPL-3.0**.
+  See ``kog2p/LICENSE.md``.
+
+Nothing in scriptconv imports these at package import time; they load only
+when a caller explicitly requests the corresponding phonemizer, accepting the
+subpackage's license by doing so.  Unencumbered alternatives exist for both
+(arbtok for Arabic, g2pk for Korean) and remain the defaults.
+"""

scriptconv-0.0.3a22/scriptconv/phonemizers/_vendored/kog2p/LICENSE.md

@@ -0,0 +1,5 @@
+# Third-party license notice — KoG2P
+
+Vendored from https://github.com/scarletcho/KoG2P, licensed **GPL-3.0** —
+NOT the Apache-2.0 license of the rest of this repository. Loaded only on an
+explicit `Phonemizer.KOG2PK` request; the Korean default is g2pk.

scriptconv-0.0.3a22/scriptconv/phonemizers/_vendored/kog2p/__init__.py

@@ -0,0 +1,383 @@
+# taken from https://github.com/scarletcho/KoG2P
+'''
+g2p.py
+~~~~~~~~~~
+
+This script converts Korean graphemes to romanized phones and then to pronunciation.
+
+    (1) graph2phone: convert Korean graphemes to romanized phones
+    (2) phone2prono: convert romanized phones to pronunciation
+    (3) graph2phone: convert Korean graphemes to pronunciation
+
+Usage:  $ python g2p.py '스물 여덟째 사람'
+        (NB. Please check 'rulebook_path' before usage.)
+
+Yejin Cho (ycho@utexas.edu)
+Jaegu Kang (jaekoo.jk@gmail.com)
+Hyungwon Yang (hyung8758@gmail.com)
+Yeonjung Hong (yvonne.yj.hong@gmail.com)
+
+Created: 2016-08-11
+Last updated: 2019-01-31 Yejin Cho
+
+* Key updates made:
+    - Executable in both Python 2 and 3.
+    - G2P Performance test available ($ python g2p.py test)
+    - G2P verbosity control available
+
+'''
+
+import datetime as dt
+import os.path
+import re
+import math
+import sys
+import optparse
+
+# Option
+# Import-time CLI parsing removed for library use (the upstream script
+# parsed sys.argv on import, which breaks under any host process with its
+# own arguments); verbose output is off in library context.
+verbose = False
+
+# Check Python version
+ver_info = sys.version_info
+
+if ver_info[0] == 2:
+    reload(sys)
+    sys.setdefaultencoding('utf-8')
+
+
+def readfileUTF8(fname):
+    f = open(fname, 'r')
+    corpus = []
+
+    while True:
+        line = f.readline()
+        line = line.encode("utf-8")
+        line = re.sub(u'\n', u'', line)
+        if line != u'':
+            corpus.append(line)
+        if not line: break
+
+    f.close()
+    return corpus
+
+
+def writefile(body, fname):
+    out = open(fname, 'w')
+    for line in body:
+        out.write('{}\n'.format(line))
+    out.close()
+
+
+def readRules(pver, rule_book):
+    if pver == 2:
+        f = open(rule_book, 'r')
+    elif pver == 3:
+        f = open(rule_book, 'r', encoding="utf-8")
+
+    rule_in = []
+    rule_out = []
+
+    while True:
+        line = f.readline()
+        if pver == 2:
+            line = unicode(line.encode("utf-8"))
+            line = re.sub(u'\n', u'', line)
+        elif pver == 3:
+            line = re.sub('\n', '', line)
+
+        if line != u'':
+            if line[0] != u'#':
+                IOlist = line.split('\t')
+                rule_in.append(IOlist[0])
+                if IOlist[1]:
+                    rule_out.append(IOlist[1])
+                else:  # If output is empty (i.e. deletion rule)
+                    rule_out.append(u'')
+        if not line: break
+    f.close()
+
+    return rule_in, rule_out
+
+
+def isHangul(charint):
+    hangul_init = 44032
+    hangul_fin = 55203
+    return charint >= hangul_init and charint <= hangul_fin
+
+
+def checkCharType(var_list):
+    #  1: whitespace
+    #  0: hangul
+    # -1: non-hangul
+    checked = []
+    for i in range(len(var_list)):
+        if var_list[i] == 32:  # whitespace
+            checked.append(1)
+        elif isHangul(var_list[i]):  # Hangul character
+            checked.append(0)
+        else:  # Non-hangul character
+            checked.append(-1)
+    return checked
+
+
+def graph2phone(graphs):
+    # Encode graphemes as utf8
+    try:
+        graphs = graphs.decode('utf8')
+    except AttributeError:
+        pass
+
+    integers = []
+    for i in range(len(graphs)):
+        integers.append(ord(graphs[i]))
+
+    # Romanization (according to Korean Spontaneous Speech corpus; 성인자유발화코퍼스)
+    phones = ''
+    ONS = ['k0', 'kk', 'nn', 't0', 'tt', 'rr', 'mm', 'p0', 'pp',
+           's0', 'ss', 'oh', 'c0', 'cc', 'ch', 'kh', 'th', 'ph', 'h0']
+    NUC = ['aa', 'qq', 'ya', 'yq', 'vv', 'ee', 'yv', 'ye', 'oo', 'wa',
+           'wq', 'wo', 'yo', 'uu', 'wv', 'we', 'wi', 'yu', 'xx', 'xi', 'ii']
+    COD = ['', 'kf', 'kk', 'ks', 'nf', 'nc', 'nh', 'tf',
+           'll', 'lk', 'lm', 'lb', 'ls', 'lt', 'lp', 'lh',
+           'mf', 'pf', 'ps', 's0', 'ss', 'oh', 'c0', 'ch',
+           'kh', 'th', 'ph', 'h0']
+
+    # Pronunciation
+    idx = checkCharType(integers)
+    iElement = 0
+    while iElement < len(integers):
+        if idx[iElement] == 0:  # not space characters
+            base = 44032
+            df = int(integers[iElement]) - base
+            iONS = int(math.floor(df / 588)) + 1
+            iNUC = int(math.floor((df % 588) / 28)) + 1
+            iCOD = int((df % 588) % 28) + 1
+
+            s1 = '-' + ONS[iONS - 1]  # onset
+            s2 = NUC[iNUC - 1]  # nucleus
+
+            if COD[iCOD - 1]:  # coda
+                s3 = COD[iCOD - 1]
+            else:
+                s3 = ''
+            tmp = s1 + s2 + s3
+            phones = phones + tmp
+
+        elif idx[iElement] == 1:  # space character
+            tmp = '#'
+            phones = phones + tmp
+
+        phones = re.sub('-(oh)', '-', phones)
+        iElement += 1
+        tmp = ''
+
+    # 초성 이응 삭제
+    phones = re.sub('^oh', '', phones)
+    phones = re.sub('-(oh)', '', phones)
+
+    # 받침 이응 'ng'으로 처리 (Velar nasal in coda position)
+    phones = re.sub('oh-', 'ng-', phones)
+    phones = re.sub('oh([# ]|$)', 'ng', phones)
+
+    # Remove all characters except Hangul and syllable delimiter (hyphen; '-')
+    phones = re.sub(r'(\W+)\-', '\\1', phones)
+    phones = re.sub(r'\W+$', '', phones)
+    phones = re.sub(r'^\-', '', phones)
+    return phones
+
+
+def phone2prono(phones, rule_in, rule_out):
+    # Apply g2p rules
+    for pattern, replacement in zip(rule_in, rule_out):
+        # print pattern
+        phones = re.sub(pattern, replacement, phones)
+        prono = phones
+    return prono
+
+
+def addPhoneBoundary(phones):
+    # Add a comma (,) after every second alphabets to mark phone boundaries
+    ipos = 0
+    newphones = ''
+    while ipos + 2 <= len(phones):
+        if phones[ipos] == u'-':
+            newphones = newphones + phones[ipos]
+            ipos += 1
+        elif phones[ipos] == u' ':
+            ipos += 1
+        elif phones[ipos] == u'#':
+            newphones = newphones + phones[ipos]
+            ipos += 1
+
+        newphones = newphones + phones[ipos] + phones[ipos + 1] + u','
+        ipos += 2
+
+    return newphones
+
+
+def addSpace(phones):
+    ipos = 0
+    newphones = ''
+    while ipos < len(phones):
+        if ipos == 0:
+            newphones = newphones + phones[ipos] + phones[ipos + 1]
+        else:
+            newphones = newphones + ' ' + phones[ipos] + phones[ipos + 1]
+        ipos += 2
+
+    return newphones
+
+
+def graph2prono(graphs, rule_in, rule_out):
+    romanized = graph2phone(graphs)
+    romanized_bd = addPhoneBoundary(romanized)
+    prono = phone2prono(romanized_bd, rule_in, rule_out)
+
+    prono = re.sub(u',', u' ', prono)
+    prono = re.sub(u' $', u'', prono)
+    prono = re.sub(u'#', u'-', prono)
+    prono = re.sub(u'-+', u'-', prono)
+
+    prono_prev = prono
+    identical = False
+    loop_cnt = 1
+
+    if verbose == True:
+        print('=> Romanized: ' + romanized)
+        print('=> Romanized with boundaries: ' + romanized_bd)
+        print('=> Initial output: ' + prono)
+
+    while not identical:
+        prono_new = phone2prono(re.sub(u' ', u',', prono_prev + u','), rule_in, rule_out)
+        prono_new = re.sub(u',', u' ', prono_new)
+        prono_new = re.sub(u' $', u'', prono_new)
+
+        if re.sub(u'-', u'', prono_prev) == re.sub(u'-', u'', prono_new):
+            identical = True
+            prono_new = re.sub(u'-', u'', prono_new)
+            if verbose == True:
+                print('\n=> Exhaustive rule application completed!')
+                print('=> Total loop count: ' + str(loop_cnt))
+                print('=> Output: ' + prono_new)
+        else:
+            if verbose == True:
+                print('\n=> Rule applied for more than once')
+                print('cmp1: ' + re.sub(u'-', u'', prono_prev))
+                print('cmp2: ' + re.sub(u'-', u'', prono_new))
+            loop_cnt += 1
+            prono_prev = prono_new
+
+    return prono_new
+
+
+def testG2P(rulebook, testset):
+    [testin, testout] = readRules(ver_info[0], testset)
+    cnt = 0
+    body = []
+    for idx in range(0, len(testin)):
+        print('Test item #: ' + str(idx + 1) + '/' + str(len(testin)))
+        item_in = testin[idx]
+        item_out = testout[idx]
+        ans = graph2phone(item_out)
+        ans = re.sub(u'-', u'', ans)
+        ans = addSpace(ans)
+
+        [rule_in, rule_out] = readRules(ver_info[0], rulebook)
+        pred = graph2prono(item_in, rule_in, rule_out)
+
+        if pred != ans:
+            print('G2P ERROR:  [result] ' + pred + '\t\t\t[ans] ' + item_in + ' [' + item_out + '] ' + ans)
+            cnt += 1
+        else:
+            body.append('[result] ' + pred + '\t\t\t[ans] ' + item_in + ' [' + item_out + '] ' + ans)
+
+    print('Total error item #: ' + str(cnt))
+    writefile(body, 'good.txt')
+
+
+kog2p_to_hangul = {
+    # Onset consonants
+    'p0': 'ㅂ',
+    'ph': 'ㅍ',
+    'pp': 'ㅃ',
+    't0': 'ㄷ',
+    'th': 'ㅌ',
+    'tt': 'ㄸ',
+    'k0': 'ㄱ',
+    'kh': 'ㅋ',
+    'kk': 'ㄲ',
+    's0': 'ㅅ',
+    'ss': 'ㅆ',
+    'h0': 'ㅎ',
+    'c0': 'ㅈ',
+    'ch': 'ㅊ',
+    'cc': 'ㅉ',
+    'mm': 'ㅁ',
+    'nn': 'ㄴ',
+    'rr': 'ㄹ',
+
+    # Coda consonants
+    'pf': 'ㅂ',
+    'tf': 'ㄷ',
+    'kf': 'ㄱ',
+    'mf': 'ㅁ',
+    'nf': 'ㄴ',
+    'ng': 'ㅇ',
+    'll': 'ㄹ',
+    'ks': 'ㄱㅅ',
+    'nc': 'ㄴㅈ',
+    'nh': 'ㄴㅎ',
+    'lk': 'ㄹㄱ',
+    'lm': 'ㄹㅁ',
+    'lb': 'ㄹㅂ',
+    'ls': 'ㄹㅅ',
+    'lt': 'ㄹㅌ',
+    'lp': 'ㄹㅍ',
+    'lh': 'ㄹㅎ',
+    'ps': 'ㅂㅅ',
+
+    # Monophthongs (vowels)
+    'ii': 'ㅣ',
+    'ee': 'ㅔ',
+    'qq': 'ㅐ',
+    'aa': 'ㅏ',
+    'xx': 'ㅡ',
+    'vv': 'ㅓ',
+    'uu': 'ㅜ',
+    'oo': 'ㅗ',
+
+    # Diphthongs (vowels)
+    'ye': 'ㅖ',
+    'yq': 'ㅒ',
+    'ya': 'ㅑ',
+    'yv': 'ㅕ',
+    'yu': 'ㅠ',
+    'yo': 'ㅛ',
+    'wi': 'ㅟ',
+    'wo': 'ㅚ',
+    'wq': 'ㅙ',
+    'we': 'ㅞ',
+    'wa': 'ㅘ',
+    'wv': 'ㅝ',
+    'xi': 'ㅢ',
+}
+
+
+def runKoG2P(graph, rulebook=None):
+    if not rulebook:
+        rulebook = f"{os.path.dirname(__file__)}/rulebook.txt"
+    [rule_in, rule_out] = readRules(ver_info[0], rulebook)
+    words = graph.split()
+    phonemized = []
+    for w in words:
+        phones = graph2prono(w, rule_in, rule_out).split()
+        hangul = "".join([kog2p_to_hangul[p] for p in phones])
+        phonemized.append(hangul)
+    return " ".join(phonemized)
+
+
+
+

scriptconv-0.0.3a22/scriptconv/phonemizers/_vendored/kog2p/rulebook.txt

@@ -0,0 +1,212 @@
+### ----------------------------------------------------------------------------------------
+### 한국어 발음규칙 (Korean G2P Rulebook)
+### Last updated: 2019-01-31
+### Yejin Cho (ycho@utexas.edu)
+### 처리순서:
+###		예외처리 - 유기음화(겹받침) - 겹받침관련규칙 - 경음화 - 겹받침단순화 - 비음화 - 리을 재음절화
+###		- 유음화 - 구개음화 - 유기음화(홑받침) - 연음 - 종성중화 - 리을 재음절화 [종료]
+### ----------------------------------------------------------------------------------------
+### 예외처리
+ii,ll,[#-]y([aeoquv]),	ii,ll,rr,y\1,	# 일 연대, 삼십일여간
+(h0,aa|t0,xx),ll,-ii,ll,	\1,ll,rr,ii,ll,	# 들일, 볼일, 할일
+s0,vv,ll,-ii,kf,	s0,vv,ll,rr,ii,kf,	# 설익(다)
+mm,uu,ll,-k0,oo,-k0,ii,	mm,uu,ll,kk,oo,k0,ii,	# 물고기
+s0,ii,ll,-s0,ii,ll	s0,ii,ll,s0,ii,ll	# 실실
+k0,ii,-s0,xx,lk,	k0,ii,s0,xx,kf,	# 기슭
+c0,vv,ll,-ya,kf,	c0,vv,rr,ya,kf,	# 절약
+k0,xx,mf,-yo,-ii,ll,	k0,xx,-mm,yo,-ii,ll,	# 금요일
+lt,-ii,	ll,-ch,ii,	# 훑이
+(?<=nn,vv,)lb,(?=-(c0,(uu|vv),kf|t0,(uu|vv),ng))	pf,	# 넓죽/넓둥글다
+(?<=s0,ii,)lh,-c0,(?=xx,ng)	ll,cc,	# 싫증
+t0,aa,lk,	t0,aa,kf,	# 닭
+(wq|we|oo),nf,-k0,aa,c0,	\1,nf,k0,aa,tf,	# 온갖
+mm,aa,tf,-h0,yv,ng,	mm,aa,th,yv,ng,	# 맏형
+k0,vv,th,-oo,s0,	k0,vv,t0,oo,tf,	# 겉옷
+c0,uu,ll,-nn,vv,mf,-k0,ii,	c0,uu,ll,rr,vv,mf,-kk,ii,	# 줄넘기
+h0,oo,th,-ii,-p0,uu,ll,	h0,oo,nf,nn,ii,p0,uu,ll,	# 홑이불
+s0,aa,ks,-ii,ll,	s0,aa,ng,nn,ii,ll,	# 삯일
+mm,qq,nf,-ii,pf,	mm,qq,nf,nn,ii,pf,	# 맨입
+kk,oo,ch,-ii,ph,	kk,oo,nf,nn,ii,pf,	# 꽃잎
+nn,qq,-p0,oo,kf,-ya,kf,	nn,qq,p0,oo,ng,nn,ya,kf,	# 내복약
+h0,aa,nf,-yv,-rr,xx,mf,	h0,aa,nf,nn,yv,rr,xx,mf,	# 한여름
+nn,aa,mf,-c0,oo,nf,-yv,-p0,ii,	nn,aa,mf,c0,oo,nf,nn,yv,p0,ii,	# 남존여비
+s0,ii,nf,-yv,-s0,vv,ng,	s0,ii,nf,nn,yv,s0,vv,ng,	# 신여성
+s0,qq,kf,-yv,nf,-ph,ii,ll,	s0,qq,ng,nn,yv,nf,ph,ii,ll,	# 색연필
+t0,aa,mf,-yo,	t0,aa,mf,nn,yo,	# 담요
+nn,uu,nf,-yo,-k0,ii,	nn,uu,nf,nn,yo,k0,ii,	# 눈요기
+vv,pf,-yo,ng,	vv,mf,nn,yo,ng,	# (영)업용
+s0,ii,kf,-yo,ng,-yu,	s0,ii,k0,yo,ng,nn,yu,	# 식용유
+nf,-yu,nf,-rr,ii,	nf,nn,yu,ll,rr,ii,	# (국민)윤리
+(c0|s0),(aa|oo|uu),ll,-ii,(ph|p0|pf),	\1,\2,ll,rr,ii,pf,	# 잘입다, 솔잎, 술잎
+(?=(^|#))h0,aa,nf,-ii,ll,	h0,aa,nf,nn,ii,ll,	# 한일
+(?=(^|#))mm,aa,kf,-ii,ll,	mm,aa,ng,nn,ii,ll,	# 막일
+mm,oo,ll,-s0,aa,ng,-s0,ii,kf,	mm,oo,ll,ss,aa,ng,s0,ii,kf,	# 몰상식
+oo,s0,#ii,pf,	oo,nf,nn,ii,pf,	# 옷입(다)
+(nf|ll),-yv,-s0,vv,-s0,	\1,nn,yv,s0,vv,tf,	# (스물/서른)여섯
+(ng|mf|nf),-y([aeoquv]),	\1,nn,y\2,	# 밤윷, 직행열차, 콩엿, 볶은엿
+(wv|ii),ll,-y([aeoquv]),	\1,rr,y\2,	# 일/월요일
+ll,-y([aeoquv]),	ll,rr,y\1,	# 불여우, 물약, 서울역, 물엿, 물옆, 굴옆, 휘발유, 유들유들
+ii,ll,-c0,vv,ll,	ii,ll,cc,vv,ll,	# 일절
+(th|tf|s0),-y([aeoquv]),	nf,-nn,y\2,	# 쑥갓요
+(<=^|#)mm,aa,kf,-ii,ll	mm,aa,ng,nn,ii,ll	# 막일
+k0,uu,-k0,xx,nf,-rr,yu,	k0,uu,k0,xx,nf,nn,yu,	# 구근류
+k0,aa,ll,-([ct])0,xx,ng,	k0,aa,ll,\1\1,xx,ng,	# 갈등/갈증
+p0,aa,ll,-t0,oo,ng,	p0,aa,ll,tt,oo,ng,	# 발동
+c0,vv,ll,-t0,oo,	c0,vv,ll,tt,oo,	# 절도
+mm,aa,ll,-s0,aa,ll,	mm,aa,ll,ss,aa,ll,	# 말살
+p0,uu,ll,-s0,	p0,uu,ll,ss,	# 불소/불세출
+ii,ll,-s0,ii,	ii,ll,ss,ii,	# 일시
+p0,aa,ll,-c0,vv,nf,	p0,aa,ll,cc,vv,nf,	# 발전
+(?<=(s0,ii,nf,|s0,aa,mf,)-)(c|k|t)0,	\2\2,	# 신고, 신다, 신자, 삼고, 삼다, 삼자
+(?<=k0,ii,mf,-)p0,	pp,	# 김밥
+(?<=t0,vv,-t0,xx,mf,-)c0,	cc,	# 더듬지
+c0,aa,mf,-c0,aa,-rr,ii,	c0,aa,mf,cc,aa,rr,ii,	# 잠자리
+(?<=(ng|ll),-)c0,(?=uu,ll,-k0,ii)	cc,	# 물줄기, 강줄기
+(?<=(nf|ll),-)p0,vv,pf,	pp,vv,pf,	# 문법, 불법
+(?<=(nf|tf),-)p0,(?=aa,-rr,aa,mf)	pp,	# 신바람, 늦바람
+p0,aa,-rr,aa,mf,-k0,yv,ll,	p0,aa,rr,aa,mf,kk,yv,ll,	# 바람결
+(?<=(mf|kf),-)p0,(?=aa,pf,)	pp,	# 아침밥, 점심밥, 저녁밥
+(?<=nn,uu,nf,-)t0,	tt,	# 눈동자, 눈대중
+mm,aa,kf,-yv,mf,	mm,aa,ng,nn,yv,mf,	# 늑막염, 결막염
+p0,aa,lb,-(t|k)0,	p0,aa,pf,\1\1,	# 밟다, 밟고
+p0,aa,lb,-nn,	p0,aa,mf,nn,	# 밟는
+nn,vv,lb,-(t|k)0,	nn,vv,ll,\1\1,	# 넓다, 넓고
+mm,(aa|vv),s0,-ii,ss,-t0,aa,	mm,\1,t0,ii,tf,tt,aa,	# 맛있다
+mm,(aa|vv),s0,-vv,ps,-t0,aa,	mm,\1,t0,vv,pf,tt,aa,	# 맛없다
+c0,vv,c0,-vv,-mm,ii,	c0,vv,t0,vv,mm,ii,	# 젖어미
+h0,vv,s0,-uu,s0,-xx,mf,	h0,vv,t0,uu,s0,xx,mf,	# 헛웃음
+k0,aa,ps,-vv,-ch,ii,	k0,aa,p0,vv,ch,ii,	# 값어치
+k0,aa,ps,-ii,ss,-nn,xx,nf,	k0,aa,p0,ii,nf,nn,xx,nf,	# 값있는
+c0,vv,lm,-c0,ii,	c0,vv,mf,cc,ii,	# 젊지
+oo,lm,-k0,(?=[iy])	oo,mf,k0,	# 옮기(다)
+k0,uu,lm,-k0,ii,-t0,aa,	k0,uu,mf,k0,ii,t0,aa,	# 굶기다
+(nn|k0|h0),aa,ll,-(p|s|c|k|t)0,	\1,aa,ll,\2\2,	# 갈바, 할바, 만날것
+ch,vv,s0,-ii,nf,	ch,vv,t0,ii,nf,	# 첫인(상)
+(?<=(mf|nf),-)ii,-p0,uu,ll,	nn,ii,p0,uu,ll,	# 솜이불
+(?<=(nf|ll),-)k0,oo,-rr,ii,	kk,oo,rr,ii,	# 문고리
+(?<=(nf|ll),-)s0,qq,	ss,qq,	# 산새, 들새
+(?<=(nf|ll),-)c0,qq,-c0,uu,	cc,qq,c0,uu,	# 손재주, 글재주
+k0,ii,ll,-k0,aa,	k0,ii,ll,kk,aa,	# 길가
+mm,uu,ll,-t0,oo,ng,-ii,	mm,uu,ll,tt,oo,ng,ii,	# 물동이
+mm,uu,ll,-c0,	mm,uu,ll,-cc,	# 물증
+(?<=(nf|ll),-)p0,aa,-t0,aa,kf,	pp,aa,t0,aa,kf,	# 발바닥, 손바닥
+(?<=(nf|ll),-)s0,oo,kf,	ss,oo,kf,	# 굴속, 물속
+(?<=s0,uu,ll,-)(c|p|t)0,	\1\1,	# 술잔, 술독, 술병, 술자리
+k0,aa,ng,-k0,aa,	k0,aa,ng,kk,aa,	# 강가
+(?<=(ng|mf),-)t0,aa,ll,	tt,aa,ll,	# 초승달
+t0,xx,ng,-p0,uu,ll,	t0,xx,ng,pp,uu,ll,	# 등불
+ch,aa,ng,-s0,aa,ll,	ch,aa,ng,ss,aa,ll,	# 창살
+(?<=(ll|ng),-)c0,uu,ll,-k0,ii,	k0,aa,ng,cc,uu,ll,k0,ii,	# 강줄기, 물줄기
+aa,nf,-k0,oo,	aa,nf,kk,oo,	# 안고
+(?<=kk,yv,-aa,nf,-)(t|c)0,	\1\1,	# 껴안지, 껴안다
+ii,-c0,uu,kf,-ii,-c0,uu,kf,	ii,c0,uu,ng,nn,ii,c0,uu,kf,	# 이죽이죽
+ya,-k0,xx,mf,-ya,-k0,xx,mf,	ya,k0,xx,mf,nn,ya,k0,xx,mf,	# 야금야금
+p0,ee,-k0,qq,s0,-ii,s0,	p0,ee,k0,qq,nf,nn,ii,tf,	# 베갯잇
+kk,qq,s0,-ii,ph,	kk,qq,nf,nn,ii,pf,	# 깻잎
+nn,aa,-mm,uu,s0,-ii,ph,	nn,aa,mm,uu,nf,nn,ii,pf,	# 나뭇잎
+qq,s0,-yv,ll,	qq,nf,nn,yv,ll,	# 도리깻열
+t0,wi,s0,-(?=[aeqiouyvwx])	t0,wi,nf,-nn,	# 뒷윷, 뒷얘기
+nn,xx,c0,-yv,-rr,xx,mf,	nn,xx,tf,nn,yv,rr,xx,mf,	# 늦여름
+t0,ii,-k0,xx,tf,-(ii|xx|ee),	t0,ii,k0,xx,s0,\1,	# 디귿에, 디귿이
+(c0|ch|th|h0),ii,-xx,(c0|ch|th|h0),-(ii|xx|ee),	\1,ii,xx,s0,\3,	# 치읓이, 지읒에
+ph,ii,-xx,ph,-(ii|xx|ee),	ph,ii,xx,p0,\1,	# 피읖에
+kh,ii,-xx,kh,-(ii|xx|ee),	kh,ii,xx,k0,\1,	# 키읔이
+### 유기음화 (겹받침)
+l(b|p),-h0,	ll,-ph,
+nh,-(c|k|t)0,	nf,-\1h,
+lh,-(c|k|t)0,	ll,-\1h,
+lk,-h0,	ll,-kh,
+nc,-h0,	nf,-ch,
+### 겹받침 규칙 (ㄹㅎ)
+(k0,aa,|k0,uu,|k0,vv,|oo,|p0,aa,|nn,aa,|nn,xx,|p0,uu,|^ii,|-,ii,mm,aa,|mm,uu,|(^|-,)vv,)lk,-(t0|c0|s0),	\1kf,-\3,
+(k0,aa,|k0,uu,|k0,vv,|vv,|oo,|mm,aa,|p0,aa,|nn,aa,|nn,xx,|mm,uu,|p0,uu,|^ii,|-,ii,)lk,-k0,	\1ll,-kk,
+### 겹받침 규칙 (ㄴㅎ)
+nh,-(k|t|c)0,	nf,-\1h,
+nh,-s0,	nf,-ss,
+nh,-nn,	nf,-nn,
+nh,-(?=[aeqiouyvwx])	-nn,
+### 겹받침 규칙 (ㄹㅎ)
+lh,-nn,	ll,-rr,
+lh,-(k|t|c)0,	ll,-\1h,
+lh,-s0,	ll,-ss,
+lh,-(?=[aeqiouyvwx])	-rr,
+### 겹받침 규칙 (ㄴㅈ)
+nc,-([ktsc])0,	nf,-\1\1,
+### 겹받침 규칙 (ㄹㅁ)
+(c0,vv,|c0,ii,|k0,uu,|t0,aa,|(^|-,)oo,|k0,oo,)lm,-([ktsc])0,	\1mf,-\3\3,
+### 겹받침 규칙 (ㄹㅂ)
+(p0,aa,|tt,vv,|(^|-,)yv,|nn,vv,|(^|-,)ya,|cc,aa,)lb,-([ktsc])0,	\1ll,-\4\4,
+### 겹받침 규칙 (ㄹㅌ)
+h0,(aa|uu),lt,-nn,	h0,\1,ll,-ll,
+h0,(aa|uu),lt,-([ktsc])0,	h0,\1,ll,-\2\2,
+### 경음화
+lk,-(c|k|p|s|t)0,	kf,-\1\1,
+l(b|p),-p0,	pf,-pp,
+s0,-p0,	tf,-pp,
+l(b|t),-(c|k|s|t|p)0,	ll,-\2\2,
+lp,-(c|k|s|t)0,	pf,-\1\1,
+(c[h0]|s[s0]|t[fh]),-(c|k|s|t)0,	tf,-\2\2,
+k[fhks],-(c|k|p|s|t)0,	kf,-\1\1,
+p[sfh],-(c|k|p|s|t)0,	pf,-\1\1,
+(?<=(kf|kh|ks|ss|c0|ch|tf|th),-)p0,	pp,
+h0,-s0,	-ss,
+nh,-s0,	nf,-ss,
+lh,-s0,	ll,-ss,
+### 겹받침 단순화: 어말 또는 자음 앞
+(ks|lk),(?=(#|$|-[ptkshcmnr]))	kf,
+n[ch],(?=(#|$|-[ptkshcmnr]))	nf,
+l[bsth],(?=(#|$|-[ptkshcmnr]))	ll,
+lm,(?=(#|$|-[ptkshcmnr]))	mf,
+(ps|lp),(?=(#|$|-[ptkshcmnr]))	pf,
+### 겹받침 단순화: 모음 앞
+([kp])s,-(?=[aeqiouyvwx])	\1f,-ss,
+ls,-(?=[aeqiouyvwx])	ll,-ss,
+nc,-(?=[aeqiouyvwx])	nf,-c0,
+lk,-(?=[aeqiouyvwx])	ll,-k0,
+lm,-(?=[aeqiouyvwx])	ll,-mm,
+lb,-(?=[aeqiouyvwx])	ll,-p0,
+l([tp]),-(?=[aeqiouyvwx])	ll,-\1h,
+### 비음화
+(?<=[pk])0,-rr,	f,-nn,
+(c0|ch|s0|ss|tf|nh|h0),-nn,	nf,-nn,
+nc,-(p|t|k)0,	nf,-\1\1,
+nc,(?=-[ptkshcmnr])	nf,
+lm,-k0,	mf,-kk,
+lm,(?=-[ptkshcmnr])	mf,
+k[fhks],(?=-(nn|mm),)	ng,
+lk,(?=-(nn|mm),)	ng,
+p[sfh],(?=-(nn|mm),)	mf,
+l[bp],(?=-(nn|mm),)	mf,
+(?<=(mf|ng|pf|kf),-)rr,	nn,
+(c0|ch|s0|ss|tf|nh|h0),(?=-mm,)	nf,
+### 리을 재음절화
+ll,-(?=y)	-rr,
+### 유음화
+(nf|ll),-rr,	ll,-rr,
+l[lht],-nn,	ll,-rr,
+### 구개음화
+tf,-(?=[iy])	-c0,
+th,-(?=[iy])	-ch,
+tf,-h0,(?=[iy])	-ch,
+### 유기음화 (홑받침)
+(p|k)f,-h0,	-\1h,
+h0,-(c|k|t)0,	-\1h,
+(tf|th|s0),(-|#)h0,	-th,
+### 연음규칙
+(s0|ss|kk|p0|ph|pp|t0|th|tt|c0|ch|kh|kk|k0|mm|nn),-(?=[aeqiouyvwx])	-\1,
+nh,-(?=[aeqiouyvwx])	-nn,
+(s0|ss|c0|ch|th),(?=-[ptkshcmnr])	tf,
+h0,-(?=[aeqiouyvwx])	-
+lh,-?(?=[aeqiouyvwx])	-rr,
+(p|t|k)f,-?(?=[aeqiouyvwx])	-\g<1>0,
+(m|n)f,-?(?=[aeqiouyvwx])	-\1\1,
+### 종성규칙
+(s0|ss|c0|ch|th),(?=-|#|$)	tf,
+(kh|kk|ks|lk),(?=-|#|$|[ptkshcmnr])	kf,	# (ks|lk),(?=-[ptkshcmnr])	kf,
+(ph|lp|ps),(?=-|#|$|[ptkshcmnr])	pf,
+(?<=[ptkshcmnr].),-(?=[aeqiouyvwx])	,
+l[bhstp],(?=-|#|$|[ptkshcmnr])	ll,	# l[bt],(?=-[ptkshcmnr])	ll,
+nh,(?=-|#|$|[ptkshcmnr])	nf,-
+### 리을 재음절화
+(?<=[aeqiouyvwx].,)ll,-(?=[aeqiouyvwx])	-rr,
+ll,-ll,	ll,-rr,

scriptconv-0.0.3a22/scriptconv/phonemizers/_vendored/mantoq/LICENSE.md

@@ -0,0 +1,15 @@
+# Third-party license notice — mantoq
+
+This directory vendors the mantoq Arabic G2P pipeline. Its phonetisation core
+(`buck/phonetise_buckwalter.py`) is adapted from Nawar Halabi's
+Arabic-Phonetiser and is licensed under the
+**Creative Commons Attribution-NonCommercial 4.0 International License**
+(https://creativecommons.org/licenses/by-nc/4.0/) — NOT the Apache-2.0 license
+of the rest of this repository.
+
+It is kept solely for compatibility with published models trained on mantoq
+phoneme sequences, is never selected automatically, and is only constructed
+when a voice or caller explicitly requests `PhonemeType.MANTOQ`. The Arabic
+default is arbtok. Arabic ↔ Buckwalter transliteration itself is NOT part of
+this notice: it is delegated to `scriptconv.notation`, an independent
+Apache-2.0 implementation of Buckwalter's published mapping.