fix tokenization: sorted vocab + phonemizer field
Browse files- coqui-fa-custom-glow-tts/config.json +134 -134
coqui-fa-custom-glow-tts/config.json
CHANGED
|
@@ -23,141 +23,141 @@
|
|
| 23 |
"<PAD>": 0,
|
| 24 |
"<EOS>": 1,
|
| 25 |
"<BOS>": 2,
|
| 26 |
-
"
|
| 27 |
-
"
|
| 28 |
-
"
|
| 29 |
-
"
|
| 30 |
-
"
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"
|
| 37 |
-
"
|
| 38 |
-
"
|
| 39 |
-
"
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
-
"
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"
|
| 46 |
-
"
|
| 47 |
-
"
|
| 48 |
-
"
|
| 49 |
-
"
|
| 50 |
-
"
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"
|
| 67 |
-
"
|
| 68 |
-
"
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
-
"
|
| 72 |
-
"
|
| 73 |
-
"
|
| 74 |
-
"
|
| 75 |
-
"
|
| 76 |
-
"
|
| 77 |
-
"
|
| 78 |
-
"
|
| 79 |
-
"
|
| 80 |
-
"
|
| 81 |
-
"
|
| 82 |
-
"
|
| 83 |
-
"
|
| 84 |
-
"
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"
|
| 88 |
-
"
|
| 89 |
-
"
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"
|
| 93 |
-
"
|
| 94 |
-
"
|
| 95 |
"w": 72,
|
| 96 |
-
"
|
| 97 |
-
"
|
| 98 |
-
"
|
| 99 |
-
"
|
| 100 |
-
"
|
| 101 |
-
"
|
| 102 |
-
"
|
| 103 |
-
"
|
| 104 |
-
"
|
| 105 |
-
"
|
| 106 |
-
"
|
| 107 |
-
"
|
| 108 |
-
"
|
| 109 |
-
"
|
| 110 |
-
"
|
| 111 |
-
"
|
| 112 |
-
"
|
| 113 |
-
"
|
| 114 |
-
"
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"
|
| 120 |
-
"
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"
|
| 124 |
-
"
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
-
"
|
| 128 |
-
"
|
| 129 |
-
"
|
| 130 |
-
"
|
| 131 |
-
"
|
| 132 |
-
"
|
| 133 |
-
"
|
| 134 |
-
"
|
| 135 |
-
"
|
| 136 |
-
"
|
| 137 |
-
"
|
| 138 |
-
"
|
| 139 |
-
"
|
| 140 |
-
"
|
| 141 |
-
"
|
| 142 |
-
"
|
| 143 |
-
"
|
| 144 |
-
"
|
| 145 |
-
"
|
| 146 |
-
"
|
| 147 |
-
"
|
| 148 |
-
"
|
| 149 |
-
"
|
| 150 |
-
"
|
| 151 |
-
"
|
| 152 |
-
"
|
| 153 |
-
"
|
| 154 |
-
"
|
| 155 |
-
"
|
| 156 |
-
"
|
| 157 |
-
"
|
| 158 |
-
"
|
| 159 |
-
"
|
| 160 |
-
"
|
| 161 |
"!": 138,
|
| 162 |
"(": 139,
|
| 163 |
")": 140,
|
|
|
|
| 23 |
"<PAD>": 0,
|
| 24 |
"<EOS>": 1,
|
| 25 |
"<BOS>": 2,
|
| 26 |
+
"\"": 3,
|
| 27 |
+
"#": 4,
|
| 28 |
+
"$": 5,
|
| 29 |
+
"%": 6,
|
| 30 |
+
"*": 7,
|
| 31 |
+
"+": 8,
|
| 32 |
+
"/": 9,
|
| 33 |
+
"0": 10,
|
| 34 |
+
"1": 11,
|
| 35 |
+
"2": 12,
|
| 36 |
+
"3": 13,
|
| 37 |
+
"4": 14,
|
| 38 |
+
"5": 15,
|
| 39 |
+
"6": 16,
|
| 40 |
+
"7": 17,
|
| 41 |
+
"8": 18,
|
| 42 |
+
"9": 19,
|
| 43 |
+
"=": 20,
|
| 44 |
+
"A": 21,
|
| 45 |
+
"B": 22,
|
| 46 |
+
"C": 23,
|
| 47 |
+
"D": 24,
|
| 48 |
+
"E": 25,
|
| 49 |
+
"F": 26,
|
| 50 |
+
"G": 27,
|
| 51 |
+
"H": 28,
|
| 52 |
+
"I": 29,
|
| 53 |
+
"J": 30,
|
| 54 |
+
"K": 31,
|
| 55 |
+
"L": 32,
|
| 56 |
+
"M": 33,
|
| 57 |
+
"N": 34,
|
| 58 |
+
"O": 35,
|
| 59 |
+
"P": 36,
|
| 60 |
+
"R": 37,
|
| 61 |
+
"S": 38,
|
| 62 |
+
"T": 39,
|
| 63 |
+
"U": 40,
|
| 64 |
+
"V": 41,
|
| 65 |
+
"W": 42,
|
| 66 |
+
"X": 43,
|
| 67 |
+
"Y": 44,
|
| 68 |
+
"Z": 45,
|
| 69 |
+
"[": 46,
|
| 70 |
+
"]": 47,
|
| 71 |
+
"^": 48,
|
| 72 |
+
"_": 49,
|
| 73 |
+
"a": 50,
|
| 74 |
+
"b": 51,
|
| 75 |
+
"c": 52,
|
| 76 |
+
"d": 53,
|
| 77 |
+
"e": 54,
|
| 78 |
+
"f": 55,
|
| 79 |
+
"g": 56,
|
| 80 |
+
"h": 57,
|
| 81 |
+
"i": 58,
|
| 82 |
+
"j": 59,
|
| 83 |
+
"k": 60,
|
| 84 |
+
"l": 61,
|
| 85 |
+
"m": 62,
|
| 86 |
+
"n": 63,
|
| 87 |
+
"o": 64,
|
| 88 |
+
"p": 65,
|
| 89 |
+
"q": 66,
|
| 90 |
+
"r": 67,
|
| 91 |
+
"s": 68,
|
| 92 |
+
"t": 69,
|
| 93 |
+
"u": 70,
|
| 94 |
+
"v": 71,
|
| 95 |
"w": 72,
|
| 96 |
+
"x": 73,
|
| 97 |
+
"y": 74,
|
| 98 |
+
"z": 75,
|
| 99 |
+
"{": 76,
|
| 100 |
+
"}": 77,
|
| 101 |
+
"æ": 78,
|
| 102 |
+
"ç": 79,
|
| 103 |
+
"ð": 80,
|
| 104 |
+
"ħ": 81,
|
| 105 |
+
"ŋ": 82,
|
| 106 |
+
"ɑ": 83,
|
| 107 |
+
"ɔ": 84,
|
| 108 |
+
"ɖ": 85,
|
| 109 |
+
"ə": 86,
|
| 110 |
+
"ɚ": 87,
|
| 111 |
+
"ɛ": 88,
|
| 112 |
+
"ɝ": 89,
|
| 113 |
+
"ɟ": 90,
|
| 114 |
+
"ɡ": 91,
|
| 115 |
+
"ɢ": 92,
|
| 116 |
+
"ɣ": 93,
|
| 117 |
+
"ɦ": 94,
|
| 118 |
+
"ɨ": 95,
|
| 119 |
+
"ɪ": 96,
|
| 120 |
+
"ɬ": 97,
|
| 121 |
+
"ɭ": 98,
|
| 122 |
+
"ɮ": 99,
|
| 123 |
+
"ɰ": 100,
|
| 124 |
+
"ɱ": 101,
|
| 125 |
+
"ɲ": 102,
|
| 126 |
+
"ɳ": 103,
|
| 127 |
+
"ɴ": 104,
|
| 128 |
+
"ɸ": 105,
|
| 129 |
+
"ɹ": 106,
|
| 130 |
+
"ɻ": 107,
|
| 131 |
+
"ɽ": 108,
|
| 132 |
+
"ɾ": 109,
|
| 133 |
+
"ʀ": 110,
|
| 134 |
+
"ʁ": 111,
|
| 135 |
+
"ʂ": 112,
|
| 136 |
+
"ʃ": 113,
|
| 137 |
+
"ʈ": 114,
|
| 138 |
+
"ʉ": 115,
|
| 139 |
+
"ʊ": 116,
|
| 140 |
+
"ʋ": 117,
|
| 141 |
+
"ʌ": 118,
|
| 142 |
+
"ʍ": 119,
|
| 143 |
+
"ʎ": 120,
|
| 144 |
+
"ʐ": 121,
|
| 145 |
+
"ʒ": 122,
|
| 146 |
+
"ʔ": 123,
|
| 147 |
+
"ʕ": 124,
|
| 148 |
+
"ʙ": 125,
|
| 149 |
+
"ʝ": 126,
|
| 150 |
+
"ʟ": 127,
|
| 151 |
+
"ˈ": 128,
|
| 152 |
+
"ˌ": 129,
|
| 153 |
+
"ː": 130,
|
| 154 |
+
"ˑ": 131,
|
| 155 |
+
"̃": 132,
|
| 156 |
+
"̩": 133,
|
| 157 |
+
"β": 134,
|
| 158 |
+
"θ": 135,
|
| 159 |
+
"χ": 136,
|
| 160 |
+
"ⱱ": 137,
|
| 161 |
"!": 138,
|
| 162 |
"(": 139,
|
| 163 |
")": 140,
|