ASR_tokenizer / vocab.json
hiba2's picture
Upload processor
618e16d verified
{
"!": 1,
"\"": 2,
"%": 3,
"+": 4,
",": 5,
"-": 6,
".": 7,
"/": 8,
"0": 9,
"1": 10,
"2": 11,
"3": 12,
"4": 13,
"5": 14,
"6": 15,
"7": 16,
"8": 17,
"9": 18,
":": 19,
"[PAD]": 109,
"[UNK]": 108,
"a": 20,
"b": 21,
"c": 22,
"d": 23,
"e": 24,
"f": 25,
"g": 26,
"h": 27,
"i": 28,
"j": 29,
"k": 30,
"l": 31,
"m": 32,
"n": 33,
"o": 34,
"p": 35,
"r": 36,
"s": 37,
"t": 38,
"u": 39,
"v": 40,
"w": 41,
"x": 42,
"y": 43,
"z": 44,
"|": 0,
"½": 45,
"¾": 46,
"×": 47,
"õ": 48,
"،": 49,
"؛": 50,
"ء": 51,
"آ": 52,
"أ": 53,
"ؤ": 54,
"إ": 55,
"ئ": 56,
"ا": 57,
"ب": 58,
"ة": 59,
"ت": 60,
"ث": 61,
"ج": 62,
"ح": 63,
"خ": 64,
"د": 65,
"ذ": 66,
"ر": 67,
"ز": 68,
"س": 69,
"ش": 70,
"ص": 71,
"ض": 72,
"ط": 73,
"ظ": 74,
"ع": 75,
"غ": 76,
"ـ": 77,
"ف": 78,
"ق": 79,
"ك": 80,
"ل": 81,
"م": 82,
"ن": 83,
"ه": 84,
"و": 85,
"ى": 86,
"ي": 87,
"ً": 88,
"ٌ": 89,
"ٍ": 90,
"َ": 91,
"ُ": 92,
"ِ": 93,
"ّ": 94,
"ْ": 95,
"٠": 96,
"١": 97,
"٢": 98,
"٣": 99,
"٥": 100,
"٦": 101,
"٧": 102,
"٨": 103,
"٩": 104,
"٪": 105,
"چ": 106,
"—": 107
}