Add "Arabic" and "Arabic Tunisian" models

.gitattributes

@@ -169,3 +169,19 @@ vosk-model-en-us-0.22/rnnlm/final.raw filter=lfs diff=lfs merge=lfs -text
 vosk-model-small-ko-0.22/graph/HCLr.fst filter=lfs diff=lfs merge=lfs -text
 vosk-model-en-us-0.22/graph/HCLG.fst filter=lfs diff=lfs merge=lfs -text
 vosk-model-en-us-0.22/rescore/G.carpa filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/am/final.mdl filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/am/tree filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/graph/HCLG.fst filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/graph/phones/align_lexicon.int filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/graph/phones/align_lexicon.txt filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/graph/words_bw.txt filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/graph/words.txt filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/ivector/final.dubm filter=lfs diff=lfs merge=lfs -text
+vosk-model-ar-mgb2-0.4/ivector/final.ie filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/am/final.mdl filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/am/tree filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/graph/Gr.fst filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/graph/HCLr.fst filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/graph/phones/align_lexicon.txt filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/ivector/final.dubm filter=lfs diff=lfs merge=lfs -text
+vosk-model-small-ar-tn-0.1-linto/ivector/final.ie filter=lfs diff=lfs merge=lfs -text

vosk-model-ar-mgb2-0.4/README

@@ -0,0 +1,10 @@
+Arabic model trained from MGB-2 dataset
+
+Get the model here https://kaldi-asr.org/models/m9
+
+SIZE 617M
+DATE 2020-02-26
+UPLOADER Dongji Gao
+RECIP Eegs/mgb2_arabic/s5
+MODEL TYPE Chain (TDNN and LSTM)
+ERROR RATE 16.40% WER (on dev set)

vosk-model-ar-mgb2-0.4/am/final.mdl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:270d47d60692aedb6b78b913ae24ab636264e67970b0b5ffa8cee666070501b1
+size 147954790

vosk-model-ar-mgb2-0.4/am/tree

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18f4a0198af15f21433b408c25f9adc00ba310e97352256bfce34185028765e4
+size 724536

vosk-model-ar-mgb2-0.4/conf/mfcc.conf

@@ -0,0 +1,10 @@
+# config for high-resolution MFCC features, intended for neural network training.
+# Note: we keep all cepstra, so it has the same info as filterbank features,
+# but MFCC is more easily compressible (because less correlated) which is why
+# we prefer this method.
+--use-energy=false   # use average of log energy, not energy.
+--sample-frequency=16000 
+--num-mel-bins=40    
+--num-ceps=40   
+--low-freq=40    # low cutoff frequency for mel bins
+--high-freq=-200 # high cutoff frequently, relative to Nyquist of 4000 (=3800)

vosk-model-ar-mgb2-0.4/conf/model.conf

@@ -0,0 +1,10 @@
+--min-active=200
+--max-active=7000
+--beam=13.0
+--lattice-beam=6.0
+--acoustic-scale=1.0
+--frame-subsampling-factor=3
+--endpoint.silence-phones=1:2:3:4:5
+--endpoint.rule2.min-trailing-silence=0.5
+--endpoint.rule3.min-trailing-silence=1.0
+--endpoint.rule4.min-trailing-silence=2.0

vosk-model-ar-mgb2-0.4/graph/disambig_tid.int

@@ -0,0 +1,2 @@
+22981
+22982

vosk-model-ar-mgb2-0.4/graph/num_pdfs

@@ -0,0 +1 @@
+6360

vosk-model-ar-mgb2-0.4/graph/phones.txt

@@ -0,0 +1,160 @@
+<eps> 0
+SIL 1
+SIL_B 2
+SIL_E 3
+SIL_I 4
+SIL_S 5
+$_B 6
+$_E 7
+$_I 8
+$_S 9
+&_B 10
+&_E 11
+&_I 12
+&_S 13
+'_B 14
+'_E 15
+'_I 16
+'_S 17
+<_B 18
+<_E 19
+<_I 20
+<_S 21
+>_B 22
+>_E 23
+>_I 24
+>_S 25
+A_B 26
+A_E 27
+A_I 28
+A_S 29
+D_B 30
+D_E 31
+D_I 32
+D_S 33
+E_B 34
+E_E 35
+E_I 36
+E_S 37
+H_B 38
+H_E 39
+H_I 40
+H_S 41
+S_B 42
+S_E 43
+S_I 44
+S_S 45
+T_B 46
+T_E 47
+T_I 48
+T_S 49
+V_B 50
+V_E 51
+V_I 52
+V_S 53
+Y_B 54
+Y_E 55
+Y_I 56
+Y_S 57
+Z_B 58
+Z_E 59
+Z_I 60
+Z_S 61
+a_B 62
+a_E 63
+a_I 64
+a_S 65
+b_B 66
+b_E 67
+b_I 68
+b_S 69
+d_B 70
+d_E 71
+d_I 72
+d_S 73
+f_B 74
+f_E 75
+f_I 76
+f_S 77
+g_B 78
+g_E 79
+g_I 80
+g_S 81
+h_B 82
+h_E 83
+h_I 84
+h_S 85
+j_B 86
+j_E 87
+j_I 88
+j_S 89
+k_B 90
+k_E 91
+k_I 92
+k_S 93
+l_B 94
+l_E 95
+l_I 96
+l_S 97
+m_B 98
+m_E 99
+m_I 100
+m_S 101
+n_B 102
+n_E 103
+n_I 104
+n_S 105
+p_B 106
+p_E 107
+p_I 108
+p_S 109
+q_B 110
+q_E 111
+q_I 112
+q_S 113
+r_B 114
+r_E 115
+r_I 116
+r_S 117
+s_B 118
+s_E 119
+s_I 120
+s_S 121
+t_B 122
+t_E 123
+t_I 124
+t_S 125
+v_B 126
+v_E 127
+v_I 128
+v_S 129
+w_B 130
+w_E 131
+w_I 132
+w_S 133
+x_B 134
+x_E 135
+x_I 136
+x_S 137
+y_B 138
+y_E 139
+y_I 140
+y_S 141
+z_B 142
+z_E 143
+z_I 144
+z_S 145
+{_B 146
+{_E 147
+{_I 148
+{_S 149
+|_B 150
+|_E 151
+|_I 152
+|_S 153
+}_B 154
+}_E 155
+}_I 156
+}_S 157
+#0 158
+#1 159

vosk-model-ar-mgb2-0.4/graph/phones/align_lexicon.int

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fdf7d5a0049e2722df88323c48434848c93ed1805871c1f04b9360e4c1b38f5
+size 35195793

vosk-model-ar-mgb2-0.4/graph/phones/align_lexicon.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2101700722b6e45bddabaa4918f07e0d2575636c52393008eb29e0ed9d480d1
+size 40057532

vosk-model-ar-mgb2-0.4/graph/phones/disambig.int

@@ -0,0 +1,2 @@
+158
+159

vosk-model-ar-mgb2-0.4/graph/phones/disambig.txt

@@ -0,0 +1,2 @@
+#0
+#1

vosk-model-ar-mgb2-0.4/graph/phones/optional_silence.csl

@@ -0,0 +1 @@
+1

vosk-model-ar-mgb2-0.4/graph/phones/optional_silence.int

@@ -0,0 +1 @@
+1

vosk-model-ar-mgb2-0.4/graph/phones/optional_silence.txt

@@ -0,0 +1 @@
+SIL

vosk-model-ar-mgb2-0.4/graph/phones/silence.csl

@@ -0,0 +1 @@
+1:2:3:4:5

vosk-model-ar-mgb2-0.4/graph/phones/word_boundary.int

@@ -0,0 +1,157 @@
+1 nonword
+2 begin
+3 end
+4 internal
+5 singleton
+6 begin
+7 end
+8 internal
+9 singleton
+10 begin
+11 end
+12 internal
+13 singleton
+14 begin
+15 end
+16 internal
+17 singleton
+18 begin
+19 end
+20 internal
+21 singleton
+22 begin
+23 end
+24 internal
+25 singleton
+26 begin
+27 end
+28 internal
+29 singleton
+30 begin
+31 end
+32 internal
+33 singleton
+34 begin
+35 end
+36 internal
+37 singleton
+38 begin
+39 end
+40 internal
+41 singleton
+42 begin
+43 end
+44 internal
+45 singleton
+46 begin
+47 end
+48 internal
+49 singleton
+50 begin
+51 end
+52 internal
+53 singleton
+54 begin
+55 end
+56 internal
+57 singleton
+58 begin
+59 end
+60 internal
+61 singleton
+62 begin
+63 end
+64 internal
+65 singleton
+66 begin
+67 end
+68 internal
+69 singleton
+70 begin
+71 end
+72 internal
+73 singleton
+74 begin
+75 end
+76 internal
+77 singleton
+78 begin
+79 end
+80 internal
+81 singleton
+82 begin
+83 end
+84 internal
+85 singleton
+86 begin
+87 end
+88 internal
+89 singleton
+90 begin
+91 end
+92 internal
+93 singleton
+94 begin
+95 end
+96 internal
+97 singleton
+98 begin
+99 end
+100 internal
+101 singleton
+102 begin
+103 end
+104 internal
+105 singleton
+106 begin
+107 end
+108 internal
+109 singleton
+110 begin
+111 end
+112 internal
+113 singleton
+114 begin
+115 end
+116 internal
+117 singleton
+118 begin
+119 end
+120 internal
+121 singleton
+122 begin
+123 end
+124 internal
+125 singleton
+126 begin
+127 end
+128 internal
+129 singleton
+130 begin
+131 end
+132 internal
+133 singleton
+134 begin
+135 end
+136 internal
+137 singleton
+138 begin
+139 end
+140 internal
+141 singleton
+142 begin
+143 end
+144 internal
+145 singleton
+146 begin
+147 end
+148 internal
+149 singleton
+150 begin
+151 end
+152 internal
+153 singleton
+154 begin
+155 end
+156 internal
+157 singleton

vosk-model-ar-mgb2-0.4/graph/phones/word_boundary.txt

@@ -0,0 +1,157 @@
+SIL nonword
+SIL_B begin
+SIL_E end
+SIL_I internal
+SIL_S singleton
+$_B begin
+$_E end
+$_I internal
+$_S singleton
+&_B begin
+&_E end
+&_I internal
+&_S singleton
+'_B begin
+'_E end
+'_I internal
+'_S singleton
+<_B begin
+<_E end
+<_I internal
+<_S singleton
+>_B begin
+>_E end
+>_I internal
+>_S singleton
+A_B begin
+A_E end
+A_I internal
+A_S singleton
+D_B begin
+D_E end
+D_I internal
+D_S singleton
+E_B begin
+E_E end
+E_I internal
+E_S singleton
+H_B begin
+H_E end
+H_I internal
+H_S singleton
+S_B begin
+S_E end
+S_I internal
+S_S singleton
+T_B begin
+T_E end
+T_I internal
+T_S singleton
+V_B begin
+V_E end
+V_I internal
+V_S singleton
+Y_B begin
+Y_E end
+Y_I internal
+Y_S singleton
+Z_B begin
+Z_E end
+Z_I internal
+Z_S singleton
+a_B begin
+a_E end
+a_I internal
+a_S singleton
+b_B begin
+b_E end
+b_I internal
+b_S singleton
+d_B begin
+d_E end
+d_I internal
+d_S singleton
+f_B begin
+f_E end
+f_I internal
+f_S singleton
+g_B begin
+g_E end
+g_I internal
+g_S singleton
+h_B begin
+h_E end
+h_I internal
+h_S singleton
+j_B begin
+j_E end
+j_I internal
+j_S singleton
+k_B begin
+k_E end
+k_I internal
+k_S singleton
+l_B begin
+l_E end
+l_I internal
+l_S singleton
+m_B begin
+m_E end
+m_I internal
+m_S singleton
+n_B begin
+n_E end
+n_I internal
+n_S singleton
+p_B begin
+p_E end
+p_I internal
+p_S singleton
+q_B begin
+q_E end
+q_I internal
+q_S singleton
+r_B begin
+r_E end
+r_I internal
+r_S singleton
+s_B begin
+s_E end
+s_I internal
+s_S singleton
+t_B begin
+t_E end
+t_I internal
+t_S singleton
+v_B begin
+v_E end
+v_I internal
+v_S singleton
+w_B begin
+w_E end
+w_I internal
+w_S singleton
+x_B begin
+x_E end
+x_I internal
+x_S singleton
+y_B begin
+y_E end
+y_I internal
+y_S singleton
+z_B begin
+z_E end
+z_I internal
+z_S singleton
+{_B begin
+{_E end
+{_I internal
+{_S singleton
+|_B begin
+|_E end
+|_I internal
+|_S singleton
+}_B begin
+}_E end
+}_I internal
+}_S singleton

vosk-model-ar-mgb2-0.4/graph/words.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1da082badb6c7e3cbf5c3caa6cd8b62ab6ffe65a57f4e7b92e75d1c02704553e
+size 21222619

vosk-model-ar-mgb2-0.4/graph/words_bw.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a502ad2fdfd2f4e0a7a3f104738eca331a20ef4c82d32fbb06eba4d09b376ac
+size 13907872

vosk-model-ar-mgb2-0.4/graph/words_head.txt

@@ -0,0 +1 @@
+<eps> 0

vosk-model-ar-mgb2-0.4/graph/words_tail.txt

@@ -0,0 +1,3 @@
+#0 957742
+<s> 957743
+</s> 957744

vosk-model-ar-mgb2-0.4/ivector/final.dubm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efb7f72fcf10ccb465f8970e83d811f4219fce2468ebb026a628b6894688af50
+size 168048

vosk-model-ar-mgb2-0.4/ivector/final.ie

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3835f479a636ec3730edd5c937a7a046f665f61067892ace1f38eac71617222b
+size 19757687

vosk-model-ar-mgb2-0.4/ivector/final.ie.id

@@ -0,0 +1 @@
+52508e2bd5a8af67fdcd9b272a6e3f77

vosk-model-ar-mgb2-0.4/ivector/global_cmvn.stats

@@ -0,0 +1,3 @@
+ [
+  1.18965e+11 1.942701e+10 -3.341223e+10 2.586335e+10 -4.536047e+10 -5.143079e+08 -2.817998e+10 -1.106823e+10 -5.748533e+09 -8.021131e+09 2.444721e+09 -1.185678e+10 -6.421724e+09 -5.475492e+09 -4.132559e+09 -4.982299e+09 -4.688524e+09 -1.968607e+09 -2.401957e+09 -1.374661e+09 -8.737595e+08 -4.833948e+08 -2.02156e+08 6.90596e+07 4.669136e+08 1.042838e+08 8.957011e+08 4292212 6.994046e+08 2.358506e+08 -2.035312e+08 4.944966e+08 -1.953043e+08 8.404543e+08 2.283313e+08 -3.559911e+08 2.555876e+08 -5.979873e+08 2.206491e+08 -2.063428e+08 1.212011e+09 
+  1.208634e+13 6.758905e+11 1.543773e+12 1.354227e+12 2.424682e+12 7.559103e+11 1.481059e+12 9.085413e+11 7.526449e+11 7.495913e+11 6.104847e+11 6.774222e+11 4.939513e+11 4.036599e+11 3.202782e+11 2.749835e+11 1.999791e+11 1.271602e+11 8.408587e+10 5.00519e+10 2.502974e+10 8.412481e+09 1.131555e+09 3.216713e+08 4.231433e+09 1.073782e+10 1.879558e+10 2.450884e+10 2.922538e+10 3.173832e+10 3.415615e+10 3.725646e+10 3.959967e+10 3.782093e+10 2.933737e+10 2.61524e+10 2.414329e+10 1.951272e+10 1.488638e+10 9.146851e+09 0 ]

vosk-model-ar-mgb2-0.4/ivector/online_cmvn.conf

@@ -0,0 +1 @@
+# configuration file for apply-cmvn-online, used in the script ../local/online/run_online_decoding_nnet2.sh

vosk-model-ar-mgb2-0.4/ivector/splice.conf

@@ -0,0 +1,2 @@
+--left-context=3
+--right-context=3

vosk-model-ar-mgb2-0.4/scripts/buckwalter2unicode.py

@@ -0,0 +1,454 @@
+#!/usr/bin/python
+
+# buckwalter2unicode.py - A script to convert transliterated Arabic
+#                         (using the Buckwalter system) to Unicode.
+#
+# Version 0.2 - 15th September 2004
+# 
+# Andrew Roberts (andyr [at] comp (dot) leeds [dot] ac (dot) uk)
+#
+# Project homepage: http://www.comp.leeds.ac.uk/andyr/software/
+#
+# Now, listen carefully...
+#
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+#
+
+from __future__ import print_function
+import sys, getopt, codecs, os, re
+
+# Declare a dictionary with Buckwalter's ASCII symbols as the keys, and
+# their unicode equivalents as values.
+
+buck2uni = {"'": u"\u0621", # hamza-on-the-line
+            "|": u"\u0622", # madda
+            ">": u"\u0623", # hamza-on-'alif
+            "&": u"\u0624", # hamza-on-waaw
+            "<": u"\u0625", # hamza-under-'alif
+            "}": u"\u0626", # hamza-on-yaa'
+            "A": u"\u0627", # bare 'alif
+            "b": u"\u0628", # baa'
+            "p": u"\u0629", # taa' marbuuTa
+            "t": u"\u062A", # taa'
+            "v": u"\u062B", # thaa'
+            "j": u"\u062C", # jiim
+            "H": u"\u062D", # Haa'
+            "x": u"\u062E", # khaa'
+            "d": u"\u062F", # daal
+            "*": u"\u0630", # dhaal
+            "r": u"\u0631", # raa'
+            "z": u"\u0632", # zaay
+            "s": u"\u0633", # siin
+            "$": u"\u0634", # shiin
+            "S": u"\u0635", # Saad
+            "D": u"\u0636", # Daad
+            "T": u"\u0637", # Taa'
+            "Z": u"\u0638", # Zaa' (DHaa')
+            "E": u"\u0639", # cayn
+            "g": u"\u063A", # ghayn
+            "_": u"\u0640", # taTwiil
+            "f": u"\u0641", # faa'
+            "q": u"\u0642", # qaaf
+            "k": u"\u0643", # kaaf
+            "l": u"\u0644", # laam
+            "m": u"\u0645", # miim
+            "n": u"\u0646", # nuun
+            "h": u"\u0647", # haa'
+            "w": u"\u0648", # waaw
+            "Y": u"\u0649", # 'alif maqSuura
+            "y": u"\u064A", # yaa'
+            "F": u"\u064B", # fatHatayn
+            "N": u"\u064C", # Dammatayn
+            "K": u"\u064D", # kasratayn
+            "a": u"\u064E", # fatHa
+            "u": u"\u064F", # Damma
+            "i": u"\u0650", # kasra
+            "~": u"\u0651", # shaddah
+            "o": u"\u0652", # sukuun
+            "`": u"\u0670", # dagger 'alif
+            "{": u"\u0671", # waSla
+}
+
+# For a reverse transliteration (Unicode -> Buckwalter), a dictionary
+# which is the reverse of the above buck2uni is essential.
+
+uni2buck = {}
+
+# Iterate through all the items in the buck2uni dict.
+for (key, value) in buck2uni.items():
+		# The value from buck2uni becomes a key in uni2buck, and vice
+		# versa for the keys.
+		uni2buck[value] = key
+
+# Declare some global variables...
+
+
+inFilename = ""  # Name of filename containing input.
+outFilename = "" # Name of filename to send the output
+inEnc = ""       # The text encoding of the input file
+outEnc = ""      # The text encoding for the output file
+ignoreChars = "" # If lines begin with these symbols, ignore.
+columnRange = "" # Holds columns numbers to transliterate.
+delimiter = ""   # Holds user-defined column delimiter.
+reverse = 0      # When equal to 1, perform reverse transliteration, i.e.,
+                 # Unicode -> Buckwalter.
+
+# A function to print to screen the usage details of this script.
+
+def usage():
+	print("Usage: {} -i INFILE -o OUTFILE [-g CHARS -c RANGE -d CHAR".format(sys.argv[0]))
+	print("       -r -e INPUT_ENCODING, -E OUTPUT ENCODING]")
+	print("      {} -l".format(sys.argv[0]))
+	print("      {} -h".format(sys.argv[0]))
+	print("")
+	print("  -i INFILE, --input=INFILE:")
+	print("    Path to text file to be transliterated to Unicode.")
+	print("  -o OUTFILE, --output=OUTFILE:")
+	print("    Path of file to output the newly transliterated text.")
+	print("  -e ENC, --input-encoding=ENC:")
+	print("    Specify the text encoding of the source file. Default: latin_1.")
+	print("  -E ENC, --output-encoding=ENC:")
+	print("    Specify the text encoding of the target file. Default: utf_8.")
+	print("  -g CHARS, --ignore-lines=CHARS:")
+	print("    Will not transliterate lines that start with any of the CHARS")
+	print("    given. E.g., -g #; will not alter lines starting with # or ;.")
+	print("    (May need to be -g \#\; on some platforms. See README.txt.)")
+	print("  -c RANGE, --columns=RANGE:")
+	print("    If in columns, select columns to apply transliteration. Can be")
+	print("    comma separated numbers, or a range. E.g., -c 1, -c 1-3, -c 1,3.")
+	print("  -d CHAR, --delimiter=CHAR:")
+	print("    Specify the delimiter that defines the column if using the -c")
+	print("    option above. Default is ' ' (space).")
+	print("  -r, --reverse:")
+	print("    Reverses the transliteration, i.e., Arabic to Buckwalter.")
+	print("    When used, it will change the default input encoding to utf_8 and")
+	print("    output encoding to latin_1") 
+	print("  -l, --list-encodings:")
+	print("    Displays all supported file encodings.")
+	print("  -h, --help:")
+	print("    Displays this page.")
+	print("")
+
+# A function to print to screen all the available encodings supported by
+# Python.
+
+def displayEncodings():
+	print("Codec		Aliases				Languages")
+	print("ascii		646, us-ascii 			English")
+	print("cp037 		IBM037, IBM039 			English")
+	print("cp424 		EBCDIC-CP-HE, IBM424		Hebrew")
+	print("cp437 		437, IBM437 			English")
+	print("cp500 		EBCDIC-CP-BE, EBCDIC-CP-CH, IBM500 	Western Europe")
+	print("cp737						Greek")
+	print("cp775 		IBM775				Baltic languages")
+	print("cp850 		850, IBM850 			Western Europe")
+	print("cp852 		852, IBM852 			Central and Eastern Europe")
+	print("cp855 		855, IBM855 			Bulgarian, Byelorussian, Macedonian, Russian, Serbian")
+	print("cp856 		 	 			Hebrew")
+	print("cp857 		857, IBM857 			Turkish")
+	print("cp860 		860, IBM860 			Portuguese")
+	print("cp861 		861, CP-IS, IBM861		Icelandic")
+	print("cp862 		862, IBM862 			Hebrew")
+	print("cp863 		863, IBM863 			Canadian")
+	print("cp864 		IBM864				Arabic")
+	print("cp865 		865, IBM865 			Danish, Norwegian")
+	print("cp869 		869, CP-GR, IBM869 		Greek")
+	print("cp874 	  					Thai")
+	print("cp875 	  					Greek")
+	print("cp1006 	  					Urdu")
+	print("cp1026 		ibm1026				Turkish")
+	print("cp1140 		ibm1140				Western Europe")
+	print("cp1250 		windows-1250 			Central and Eastern Europe")
+	print("cp1251 		windows-1251 			Bulgarian, Byelorussian, Macedonian, Russian, Serbian")
+	print("cp1252 		windows-1252 			Western Europe")
+	print("cp1253 		windows-1253 			Greek")
+	print("cp1254 		windows-1254 			Turkish")
+	print("cp1255 		windows-1255 			Hebrew")
+	print("cp1256 		windows-1256 			Arabic")
+	print("cp1257 		windows-1257		 	Baltic languages")
+	print("cp1258 		windows-1258		 	Vietnamese")
+	print("latin_1		iso-8859-1, iso8859-1, 8859, cp819, latin, latin1, L1	West Europe")
+	print("iso8859_2 	iso-8859-2, latin2, L2		Central and Eastern Europe")
+	print("iso8859_3 	iso-8859-3, latin3, L3		Esperanto, Maltese")
+	print("iso8859_4 	iso-8859-4, latin4, L4		Baltic languagues")
+	print("iso8859_5 	iso-8859-5, cyrillic		Bulgarian, Byelorussian, Macedonian, Russian, Serbian")
+	print("iso8859_6 	iso-8859-6, arabic		Arabic")
+	print("iso8859_7 	iso-8859-7, greek, greek8	Greek")
+	print("iso8859_8 	iso-8859-8, hebrew		Hebrew")
+	print("iso8859_9 	iso-8859-9, latin5, L5		Turkish")
+	print("iso8859_10 	iso-8859-10, latin6, L6 	Nordic languages")
+	print("iso8859_13 	iso-8859-13			Baltic languages")
+	print("iso8859_14 	iso-8859-14, latin8, L8		Celtic languages")
+	print("iso8859_15 	iso-8859-15			Western Europe")
+	print("koi8_r						Russian")
+	print("koi8_u						Ukrainian")
+	print("mac_cyrillic	maccyrillic			Bulgarian, Byelorussian, Macedonian, Russian, Serbian")
+	print("mac_greek	macgreek			Greek")
+	print("mac_iceland	maciceland			Icelandic")
+	print("mac_latin2	maclatin2, maccentraleurope	Central and Eastern Europe")
+	print("mac_roman 	macroman 			Western Europe")
+	print("mac_turkish 	macturkish 			Turkish")
+	print("utf_16 		U16, utf16 			all languages")
+	print("utf_16_be 	UTF-16BE 			all languages (BMP only)")
+	print("utf_16_le 	UTF-16LE 			all languages (BMP only)")
+	print("utf_7 		U7 				all languages")
+	print("utf_8 		U8, UTF, utf8 			all languages")
+
+def parseIgnoreString(string):
+	
+	symbols = []
+	
+	for char in string:
+		symbols.append(char)
+
+	return symbols
+
+# Begin parsing the command-line arguments...
+
+try:
+	(options, args) = getopt.getopt(sys.argv[1:], "i:o:e:E:g:c:d:rlh",
+	["input=","output=", "input-encoding=", "output-encoding=",
+	"ignore-lines=", "columns=", "delimiter=" "reverse", "list-encodings",
+	"help"])
+
+except getopt.GetoptError:
+	# print help information and exit:
+	usage()
+	sys.exit(1)
+
+# Loop over all arguments supplied by the user.
+for (x, y) in options:
+	if x in ("-h", "--help"): 
+		usage()
+		sys.exit(0)
+	
+	if x in ("-l", "--list-encodings"): 
+		displayEncodings()
+		sys.exit(0)
+		
+	if x in ("-i", "--input"): inFilename = y
+	if x in ("-o", "--output"): outFilename = y
+	if x in ("-e", "--input-encoding"): inEnc= y
+	if x in ("-E", "--output-encoding"): outEnc= y
+	if x in ("-r", "--reverse"): reverse = 1
+	if x in ("-g", "--ignore-lines"): ignoreChars = y
+	if x in ("-c", "--columns"): columnRange = y
+	if x in ("-d", "--delimiter"): 
+		delimiter = y
+		# Tabs come in off the command line from "\\t" to "\t". However,
+		# that's equivalent to "\\t" from python's point of view.
+		# Therefore replace any inputted "tabs" with proper tabs before
+		# proceeding.
+		delimiter = delimiter.replace("\\t", "\t")
+		# Do some error checking
+		if len(delimiter) > 1:
+			print("Delimeter should only be a single character. Using first character" + delimiter[0], file=sys.stderr)
+			delimiter = delimiter[0]
+		
+		if buck2uni.get(delimiter):
+			print("Invalid delimiter. \"" + delimiter + "\" is part of the Buckwalter character set.", file=sys.stderr)
+			print("This will obviously cause much confusion as a delimiter!", file=sys.stderr)
+			print("Please try again. Aborting...", file=sys.stderr)
+			sys.exit(1)
+
+# If no delimiter was set then, set the default to " " (space)
+if not delimiter:
+	delimiter = " "
+
+# If user didn't specify the encoding of the input file, then revert to
+# defaults. The defaults can depending on the direction of
+# transliteration:
+#
+# Buckwalter -> Unicode, default = latin1
+# Unicode -> Buckwalter, default = utf_8
+
+
+if not inEnc:
+	if reverse:
+		inEnc = "utf_8"
+	else:
+		inEnc = "latin_1"
+
+# Similarly, if user didn't specify the encoding of the output file,
+# then revert to defaults. The defaults can depending on the direction
+# of transliteration:
+#
+# Buckwalter -> Unicode, default = utf_8 
+# Unicode -> Buckwalter, default # = latin_1
+
+if not outEnc:
+	if reverse:
+		outEnc = "latin_1"
+	else:
+		outEnc = "utf_8"
+
+# Ok, let's get the files open!
+
+# Providing a file for output was specified...
+if outFilename:
+    try:
+		# Create a file object, set it to "write" mode using the
+		# specified output encoding.
+		outFile = codecs.open(outFilename, "w", outEnc)
+
+    except IOError as msg:
+		# A problem occurred when trying to open this file. Report to
+		# user...
+        print(msg)
+        sys.exit(1)
+
+# Script can not work without somewhere to store the transliteration.
+# Exit. 
+else:
+	print("Must specify a file to use store the output! Aborting...")
+	sys.exit(1)
+
+# Providing a file for input was specified...
+if inFilename:
+    try:
+		# Create a file object, set it to "read" mode using the
+		# specified input encoding.
+		inFile = codecs.open(inFilename, "r", inEnc)
+
+    except IOError as msg:
+		# A problem occurred when trying to open this file. Report to
+		# user...
+        print(msg)
+        sys.exit(1)
+
+# This script requires a file to read from. Exit.
+else:
+	print("Must specify a file to use as input! Aborting...")
+	sys.exit(1)
+
+def getColsFromRange(cRange):
+	
+	columns = []
+	hyphenSearch = re.compile(r'-')
+
+	rangeElements = cRange.split(",")
+
+	for i in rangeElements:
+		# If it contains a hyphen (e.g., 1-3)
+		if hyphenSearch.search(i):
+			[start, end] = i.split("-")
+			columns = columns + list(range(int(start)-1,int(end)))
+		else:
+			columns.append(int(i)-1)
+
+	return columns
+
+# This function transliterates a given string. It checks the direction
+# of the transliteration and then uses the appropriate dictionary. A
+# transliterated string is returned.
+
+def transliterate(inString, lineNumber):
+	out = ""
+
+	if columnRange:
+		columns = getColsFromRange(columnRange)
+
+		# Split the line on the delimiter
+		lineCols = inString.split(delimiter)
+
+		# Iterate over each column. If it's one of the ones in the range
+		# specified, then transliterate, otherwise just output column
+		# unchanged.
+
+		for i in range(len(lineCols)):
+			
+			# If first column, then don't prefix the delimiter
+			if i == 0:
+				if i in columns:
+					out = transliterateString(lineCols[i])
+				else :
+					out = lineCols[i]
+			else :
+				if i in columns:
+					out = out + delimiter + transliterateString(lineCols[i])
+				else :
+					out = out + delimiter + lineCols[i]
+
+	else:
+		out = transliterateString(inString)
+		
+
+
+	return out
+
+def transliterateString(inString):
+
+	out = ""
+	
+	# For normal Buckwalter -> Unicode transliteration..
+	if not reverse:
+
+		# Loop over each character in the string, inString.
+		for char in inString:
+			# Look up current char in the dictionary to get its
+			# respective value. If there is no match, e.g., chars like
+			# spaces, then just stick with the current char without any
+			# conversion.
+			out = out + buck2uni.get(char, char)
+	
+	# Same as above, just in the other direction.
+	else:
+
+		for char in inString:
+			out = out + uni2buck.get(char, char)
+
+	return out
+
+#while 1:
+#	line = inFile.readline().strip()
+#	line = line.decode(inEnc)
+#	if not line:
+#		break
+
+	# process string
+#	outFile.write(transliterate(line) + os.linesep)
+
+# Read in the lines of the input file.
+lines = inFile.readlines()
+
+currentLineNumber = 1
+# Loop over each line
+for line in lines:
+	line = line.strip()
+	try:
+		# Transliterate the current line, and then write the output to
+		# file.
+		
+		if not ignoreChars:
+			outFile.write(transliterate(line, currentLineNumber) + " " + os.linesep)
+		else:
+			if line[0] in parseIgnoreString(ignoreChars):
+				outFile.write(line + " " + os.linesep)
+			else:
+				outFile.write(transliterate(line, currentLineNumber) + " " + os.linesep)
+		
+		currentLineNumber = currentLineNumber + 1
+
+	except UnicodeError as msg:
+		# A problem when writing occurred. Report to user...
+		print(msg)
+		sys.exit(1)
+
+# All done! Better close the files used before terminating...
+inFile.close()
+outFile.close()
+
+# ... and relax! :)

vosk-model-small-ar-tn-0.1-linto/am/cmvn_opts

@@ -0,0 +1 @@
+--norm-means=false --norm-vars=false

vosk-model-small-ar-tn-0.1-linto/am/final.ie.id

@@ -0,0 +1 @@
+0084a8987dd3b241beabb01efcc32e17

vosk-model-small-ar-tn-0.1-linto/am/final.mdl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9462fea3133f2b3ef672df2cabffb749f5eedf49427bff7bedd3457a9dfd7da3
+size 77422160

vosk-model-small-ar-tn-0.1-linto/am/frame_subsampling_factor

@@ -0,0 +1 @@
+3

vosk-model-small-ar-tn-0.1-linto/am/num_jobs

@@ -0,0 +1 @@
+4

vosk-model-small-ar-tn-0.1-linto/am/phones.txt

@@ -0,0 +1,302 @@
+<eps> 0
+SIL 1
+SIL_B 2
+SIL_E 3
+SIL_I 4
+SIL_S 5
+A$_B 6
+A$_E 7
+A$_I 8
+A$_S 9
+A&_B 10
+A&_E 11
+A&_I 12
+A&_S 13
+A'_B 14
+A'_E 15
+A'_I 16
+A'_S 17
+A<_B 18
+A<_E 19
+A<_I 20
+A<_S 21
+A>_B 22
+A>_E 23
+A>_I 24
+A>_S 25
+AA_B 26
+AA_E 27
+AA_I 28
+AA_S 29
+AD_B 30
+AD_E 31
+AD_I 32
+AD_S 33
+AE_B 34
+AE_E 35
+AE_I 36
+AE_S 37
+AH_B 38
+AH_E 39
+AH_I 40
+AH_S 41
+AS_B 42
+AS_E 43
+AS_I 44
+AS_S 45
+AT_B 46
+AT_E 47
+AT_I 48
+AT_S 49
+AV_B 50
+AV_E 51
+AV_I 52
+AV_S 53
+AY_B 54
+AY_E 55
+AY_I 56
+AY_S 57
+AZ_B 58
+AZ_E 59
+AZ_I 60
+AZ_S 61
+Ab_B 62
+Ab_E 63
+Ab_I 64
+Ab_S 65
+Ad_B 66
+Ad_E 67
+Ad_I 68
+Ad_S 69
+Af_B 70
+Af_E 71
+Af_I 72
+Af_S 73
+Ag_B 74
+Ag_E 75
+Ag_I 76
+Ag_S 77
+Ah_B 78
+Ah_E 79
+Ah_I 80
+Ah_S 81
+Aj_B 82
+Aj_E 83
+Aj_I 84
+Aj_S 85
+Ak_B 86
+Ak_E 87
+Ak_I 88
+Ak_S 89
+Al_B 90
+Al_E 91
+Al_I 92
+Al_S 93
+Am_B 94
+Am_E 95
+Am_I 96
+Am_S 97
+An_B 98
+An_E 99
+An_I 100
+An_S 101
+Ap_B 102
+Ap_E 103
+Ap_I 104
+Ap_S 105
+Aq_B 106
+Aq_E 107
+Aq_I 108
+Aq_S 109
+Ar_B 110
+Ar_E 111
+Ar_I 112
+Ar_S 113
+As_B 114
+As_E 115
+As_I 116
+As_S 117
+At_B 118
+At_E 119
+At_I 120
+At_S 121
+Av_B 122
+Av_E 123
+Av_I 124
+Av_S 125
+Aw_B 126
+Aw_E 127
+Aw_I 128
+Aw_S 129
+Ax_B 130
+Ax_E 131
+Ax_I 132
+Ax_S 133
+Ay_B 134
+Ay_E 135
+Ay_I 136
+Ay_S 137
+Az_B 138
+Az_E 139
+Az_I 140
+Az_S 141
+A|_B 142
+A|_E 143
+A|_I 144
+A|_S 145
+A}_B 146
+A}_E 147
+A}_I 148
+A}_S 149
+L'_B 150
+L'_E 151
+L'_I 152
+L'_S 153
+La_B 154
+La_E 155
+La_I 156
+La_S 157
+Lb_B 158
+Lb_E 159
+Lb_I 160
+Lb_S 161
+Lc_B 162
+Lc_E 163
+Lc_I 164
+Lc_S 165
+Ld_B 166
+Ld_E 167
+Ld_I 168
+Ld_S 169
+Le_B 170
+Le_E 171
+Le_I 172
+Le_S 173
+Lf_B 174
+Lf_E 175
+Lf_I 176
+Lf_S 177
+Lg_B 178
+Lg_E 179
+Lg_I 180
+Lg_S 181
+Lh_B 182
+Lh_E 183
+Lh_I 184
+Lh_S 185
+Li_B 186
+Li_E 187
+Li_I 188
+Li_S 189
+Lj_B 190
+Lj_E 191
+Lj_I 192
+Lj_S 193
+Lk_B 194
+Lk_E 195
+Lk_I 196
+Lk_S 197
+Ll_B 198
+Ll_E 199
+Ll_I 200
+Ll_S 201
+Lm_B 202
+Lm_E 203
+Lm_I 204
+Lm_S 205
+Ln_B 206
+Ln_E 207
+Ln_I 208
+Ln_S 209
+Lo_B 210
+Lo_E 211
+Lo_I 212
+Lo_S 213
+Lp_B 214
+Lp_E 215
+Lp_I 216
+Lp_S 217
+Lq_B 218
+Lq_E 219
+Lq_I 220
+Lq_S 221
+Lr_B 222
+Lr_E 223
+Lr_I 224
+Lr_S 225
+Ls_B 226
+Ls_E 227
+Ls_I 228
+Ls_S 229
+Lt_B 230
+Lt_E 231
+Lt_I 232
+Lt_S 233
+Lu_B 234
+Lu_E 235
+Lu_I 236
+Lu_S 237
+Lv_B 238
+Lv_E 239
+Lv_I 240
+Lv_S 241
+Lw_B 242
+Lw_E 243
+Lw_I 244
+Lw_S 245
+Lx_B 246
+Lx_E 247
+Lx_I 248
+Lx_S 249
+Ly_B 250
+Ly_E 251
+Ly_I 252
+Ly_S 253
+Lz_B 254
+Lz_E 255
+Lz_I 256
+Lz_S 257
+ae_B 258
+ae_E 259
+ae_I 260
+ae_S 261
+cc_B 262
+cc_E 263
+cc_I 264
+cc_S 265
+ga_B 266
+ga_E 267
+ga_I 268
+ga_S 269
+ge_B 270
+ge_E 271
+ge_I 272
+ge_S 273
+gu_B 274
+gu_E 275
+gu_I 276
+gu_S 277
+ha_B 278
+ha_E 279
+ha_I 280
+ha_S 281
+he_B 282
+he_E 283
+he_I 284
+he_S 285
+hi_B 286
+hi_E 287
+hi_I 288
+hi_S 289
+ho_B 290
+ho_E 291
+ho_I 292
+ho_S 293
+hu_B 294
+hu_E 295
+hu_I 296
+hu_S 297
+#0 298
+#1 299
+#2 300
+#3 301

vosk-model-small-ar-tn-0.1-linto/am/tree

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:841160139eae5a74a5ebb69fa407ff8db25d2e57d7344e929f246ea05b9dfc6c
+size 658228

vosk-model-small-ar-tn-0.1-linto/conf/mfcc.conf

@@ -0,0 +1,10 @@
+# config for high-resolution MFCC features, intended for neural network training.
+# Note: we keep all cepstra, so it has the same info as filterbank features,
+# but MFCC is more easily compressible (because less correlated) which is why
+# we prefer this method.
+--use-energy=false   # use average of log energy, not energy.
+--sample-frequency=16000 
+--num-mel-bins=40    
+--num-ceps=40   
+--low-freq=40    # low cutoff frequency for mel bins
+--high-freq=-200 # high cutoff frequently, relative to Nyquist of 4000 (=3800)

vosk-model-small-ar-tn-0.1-linto/conf/model.conf

@@ -0,0 +1,10 @@
+--min-active=200
+--max-active=7000
+--beam=11.0
+--lattice-beam=6.0
+--acoustic-scale=1.0
+--frame-subsampling-factor=3
+--endpoint.silence-phones=1:2:3:4:5
+--endpoint.rule2.min-trailing-silence=0.5
+--endpoint.rule3.min-trailing-silence=1.0
+--endpoint.rule4.min-trailing-silence=2.0

vosk-model-small-ar-tn-0.1-linto/conf/splice.conf

@@ -0,0 +1,3 @@
+--left-context=3
+--right-context=3
+

vosk-model-small-ar-tn-0.1-linto/graph/disambig_tid.int

@@ -0,0 +1,4 @@
+14649
+14650
+14651
+14652

vosk-model-small-ar-tn-0.1-linto/graph/phones/align_lexicon.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ae5ebc3c3a557ec303d9922a31906a9110883b1f1d1f6a92e58ba3237905095
+size 16371230

vosk-model-small-ar-tn-0.1-linto/graph/phones/disambig.int

@@ -0,0 +1,4 @@
+298
+299
+300
+301

vosk-model-small-ar-tn-0.1-linto/graph/phones/disambig.txt

@@ -0,0 +1,4 @@
+#0
+#1
+#2
+#3

vosk-model-small-ar-tn-0.1-linto/graph/phones/optional_silence.csl

@@ -0,0 +1 @@
+1