Issue #761: add initial support for Turkmen language

CHANGELOG.md

@@ -29,6 +29,7 @@ new languages:
 *  ltg (Latgalian) -- Valdis Vitolins
 *  uk (Ukrainian) -- Valdis Vitolins
 *  qu (Quechua) -- Valdis Vitolins
+*  tk (Turkmen) -- boracasli98, Valdis Vitolins
 
 ### 1.50 - 2019-11-29
 

Makefile.am

@@ -393,6 +393,7 @@ phsource/phonemes.stamp: \
 	phsource/ph_telugu \
 	phsource/ph_tibetan \
 	phsource/ph_turkish \
+	phsource/ph_turkmen \
 	phsource/ph_urdu \
 	phsource/ph_uzbek \
 	phsource/ph_vietnam \
@@ -844,6 +845,9 @@ espeak-ng-data/ta_dict: dictsource/ta_list dictsource/ta_rules dictsource/ta_ext
 te: espeak-ng-data/te_dict
 espeak-ng-data/te_dict: dictsource/te_list dictsource/te_rules dictsource/te_extra dictsource/te_emoji
 
+tk: espeak-ng-data/tk_dict
+espeak-ng-data/tk_dict: dictsource/tk_list dictsource/tk_listx dictsource/tk_rules dictsource/tk_extra
+
 tn: espeak-ng-data/tn_dict
 espeak-ng-data/tn_dict: dictsource/tn_list dictsource/tn_rules dictsource/tn_extra
 

dictsource/tk_list

@@ -0,0 +1,31 @@
+// Letters
+_a A
+_ä &
+b  b'E
+ç  tS'E
+d  d'E
+_e E
+f  f,'E
+g  g'E
+h  h'E
+_i I
+j  dZ'E
+k  k'A
+l  'El
+m  'Em
+n  'En
+ň  'EN
+_o o
+_ö W
+p  p'E
+r  E'R
+s  T'E
+ş  S'E
+t  t'E
+_u U
+_ü Y
+w  B'E
+_y @
+ý  j'E
+z  D'E
+ž  Z'E

dictsource/tk_rules

@@ -0,0 +1,119 @@
+// This file is UTF8 encoded
+// Spelling to phoneme rules for Turkmen language
+
+// Letter groups
+.L01 a o u y
+.L02 ä e i ö ü
+
+.replace
+// Replace non-standard letters
+    ñ    ň
+    ÿ    ý
+
+.group a
+    a    A
+
+.group ä
+    ä    &
+
+.group b
+    b     b
+
+.group ç
+    ç    tS
+
+.group d
+    d    d
+
+.group e
+    e    e
+
+.group f
+    f    f,
+
+.group g
+    g    g
+    g (L01     Q"
+    g (L02     g
+    L01) g     Q"
+    L02) g     g
+
+.group h
+    h    h
+    h (L01     x
+    h (L02     h
+    L01) h     x
+    L02) h     h
+
+.group i
+    i    i
+
+.group j
+    j    dZ
+
+.group k
+    k    k
+    k (L01     q
+    k (L02     k
+    L01) k     q
+    L02) k     k
+
+.group l
+    l    l
+    l (L01     L
+    l (L02     l
+    L01) l     L
+    L02) l     l
+
+.group m
+    m    m
+
+.group n
+    n    n
+
+.group ň
+    ň   N
+
+.group o
+    o    o
+
+.group ö
+    ö     W
+
+.group p
+    p    p
+
+.group r
+    r    R
+
+.group s
+    s    T
+
+.group ş
+    ş    S
+
+.group t
+    t    t
+    ts    ts
+
+.group u
+    u    U
+
+.group ü
+    ü    Y
+    üý    Y:
+
+.group w
+    w    B
+
+.group y
+    y    @
+
+.group ý
+    ý    j
+
+.group z
+    z    D
+
+.group ž
+    ž     Z

docs/languages.md

@@ -8,7 +8,7 @@ and dialects,
 [private-use extensions](https://raw.githubusercontent.com/espeak-ng/bcp47-data/master/bcp47-extensions)
 have been used.
 
-The 115 supported languages and accents are:
+The 116 supported languages and accents are:
 
 | Family Code | Identifier        | Language Family       | Language                    | Accent/Dialect         |
 |-------------|-------------------|-----------------------|-----------------------------|------------------------|
@@ -118,6 +118,7 @@ The 115 supported languages and accents are:
 | `bnt`       | `sw`              | Bantu                 | Swahili                     |                        |
 | `gmq`       | `sv`              | North Germanic        | Swedish                     |                        |
 | `dra`       | `ta`              | Dravidian             | Tamil                       |                        |
+| `trk`       | `tk`              | Turkic                | Turkmen<sup>\[5\]</sup>     |                        |
 | `trk`       | `tt`              | Turkic                | Tatar                       |                        |
 | `dra`       | `te`              | Dravidian             | Telugu                      |                        |
 | `trk`       | `tr`              | Turkic                | Turkish                     |                        |
@@ -135,3 +136,7 @@ The 115 supported languages and accents are:
 \[3\] Currently, only Hiragana and Katakana are supported.
 
 \[4\] Only Cherokee-English Dictionary fully annotated UTF-8 pronunciations are supported. Syllabary is not supported.
+
+\[5\] With Latin alphabet.
+
+

espeak-ng-data/lang/trk/tk

@@ -0,0 +1,2 @@
+name Turkmen
+language tk

phsource/ph_turkmen

@@ -0,0 +1,12 @@
+phoneme A
+  vwl starttype #a endtype #a
+  length 200
+  FMT(vowel/aa_9)
+endphoneme
+
+phoneme &
+  vwl starttype #a endtype #a
+  flag1
+  length 200
+  FMT(vowel/ee_2)
+endphoneme

phsource/phonemes

@@ -2013,6 +2013,9 @@ phonemetable uk ru
 phonemetable az tr
 include ph_azerbaijani
 
+phonemetable tk tr
+include ph_turkmen
+
 phonemetable pt-pt pt
 include ph_portugal
 
@@ -2026,4 +2029,4 @@ phonemetable sd ur
 include ph_sindhi
 
 phonemetable uz kk
-include ph_uzbek
+include ph_uzbek

tests/language-pronunciation.test

@@ -232,6 +232,8 @@ test_phonemes te "Telu" "n'e:nu g'a:Ju t'inag,alanu m'ariju 'ala: c'e:sina: n'a:
 test_phonemes tn "Latn" "B'aTU B'oKl B'a ts'i:tswl B'a g,olUl,os-'igill_:_: ll g'o l,ekallk'ana k'a s-'iRiti_:_: ll d,itSwan'elU
 B'a_| aB'etswl g'o_| ak'an^a_:_: ll _|m,aik'ul#U
 ,m-ml B'a tSwan'ets-i g'o d,iRll'ana k'a _|m'owa_! wa B,okaul'eNgwl" "Batho botlhe ba tsetswe ba gololosegile le go lekalekana ka seriti le ditshwanelo. Ba abetswe go akanya le maikutlo, mme ba tshwanetse go direlana ka mowa wa bokaulengwe."
+test_phonemes tk "Latn" "tYRkm'&ntSE
+t'YRkm&n d'ilI" "Türkmençe, Türkmen dili"
 test_phonemes tr "Latn" "piZ'amaL@ hast'a j'a:@z SofW*'E tSabudZ'ak Jyv&nd'I
 SISlid'E byj'Yk tS'Wp j@:@nLa*'@" "Pijamalı hasta yağız şoföre çabucak güvendi. Şişli’de büyük çöp yığınları."
 test_phonemes tt "Cyrl" "b0rL'Vq keSel'&r d'& 0z'0t h'&m 'yz 0brujLAr'V h'&m xoquqLAr'V jAGVnn'An t'iN bup'Vp tuAp'Ar