truong-xuan-linh's picture
Upload processor
6f9fb6b verified
raw
history blame
26.3 kB
{
"added_tokens_decoder": {
"0": {
"content": "<s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<pad>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "</s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"3": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"4": {
"content": "▁",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"5": {
"content": "e",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"6": {
"content": "t",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"7": {
"content": "a",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"8": {
"content": "o",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"9": {
"content": "n",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"10": {
"content": "i",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"11": {
"content": "h",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"12": {
"content": "s",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"13": {
"content": "r",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"14": {
"content": "d",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"15": {
"content": "l",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"16": {
"content": "u",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"17": {
"content": "c",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"18": {
"content": "m",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"19": {
"content": "f",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"20": {
"content": "w",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"21": {
"content": "g",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"22": {
"content": "y",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"23": {
"content": ",",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"24": {
"content": "p",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"25": {
"content": "b",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"26": {
"content": ".",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"27": {
"content": "v",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"28": {
"content": "k",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"29": {
"content": "\"",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"30": {
"content": "I",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"31": {
"content": "'",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"32": {
"content": "T",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"33": {
"content": "A",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"34": {
"content": "S",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"35": {
"content": "H",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"36": {
"content": ";",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"37": {
"content": "x",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"38": {
"content": "W",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"39": {
"content": "-",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"40": {
"content": "B",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"41": {
"content": "?",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"42": {
"content": "C",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"43": {
"content": "M",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"44": {
"content": "!",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"45": {
"content": "q",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"46": {
"content": "j",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"47": {
"content": "E",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"48": {
"content": "N",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"49": {
"content": "P",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50": {
"content": "O",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"51": {
"content": "D",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"52": {
"content": "L",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"53": {
"content": "G",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"54": {
"content": "R",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"55": {
"content": "F",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"56": {
"content": "Y",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"57": {
"content": "z",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"58": {
"content": "J",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"59": {
"content": ":",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"60": {
"content": "K",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"61": {
"content": "U",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"62": {
"content": "V",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"63": {
"content": ")",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"64": {
"content": "(",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"65": {
"content": "Q",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"66": {
"content": "Z",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"67": {
"content": "]",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"68": {
"content": "[",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"69": {
"content": "X",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"70": {
"content": "—",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"71": {
"content": "/",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"72": {
"content": "æ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"73": {
"content": "é",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"74": {
"content": "{",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"75": {
"content": "}",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"76": {
"content": "ê",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"77": {
"content": "œ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"78": {
"content": "̄",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"79": {
"content": "<mask>",
"lstrip": true,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"80": {
"content": "<ctc_blank>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"81": {
"content": "ặ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"82": {
"content": "ỉ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"83": {
"content": "ổ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"84": {
"content": "ý",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"85": {
"content": "ờ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"86": {
"content": "ẵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"87": {
"content": "è",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"88": {
"content": "ũ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"89": {
"content": "ồ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"90": {
"content": "ể",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"91": {
"content": "ỗ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"92": {
"content": "ủ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"93": {
"content": "ỹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"94": {
"content": "ị",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"95": {
"content": "ẹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"96": {
"content": "ô",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"97": {
"content": "ư",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"98": {
"content": "ệ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"99": {
"content": "ề",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"100": {
"content": "ơ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"101": {
"content": "ỷ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"102": {
"content": "ữ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"103": {
"content": "á",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"104": {
"content": "ù",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"105": {
"content": "ứ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"106": {
"content": "ố",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"107": {
"content": "ẩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"108": {
"content": "â",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"109": {
"content": "ạ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"110": {
"content": "ợ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"111": {
"content": "ắ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"112": {
"content": "ấ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"113": {
"content": "ã",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"114": {
"content": "à",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"115": {
"content": "ọ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"116": {
"content": "ă",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"117": {
"content": "ĩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"118": {
"content": "ằ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"119": {
"content": "ẳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"120": {
"content": "ễ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"121": {
"content": "ó",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"122": {
"content": "ẫ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"123": {
"content": "đ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"124": {
"content": "ẽ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"125": {
"content": "ừ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"126": {
"content": "ở",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"127": {
"content": "ầ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"128": {
"content": "ì",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"129": {
"content": "ú",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"130": {
"content": "ớ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"131": {
"content": "í",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"132": {
"content": "õ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"133": {
"content": "ỵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"134": {
"content": "ộ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"135": {
"content": "ử",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"136": {
"content": "ả",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"137": {
"content": "ự",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"138": {
"content": "ỳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"139": {
"content": "ẻ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"140": {
"content": "ỏ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"141": {
"content": "ỡ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"142": {
"content": "ụ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"143": {
"content": "ậ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"144": {
"content": "ế",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"145": {
"content": "ò",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"146": {
"content": "ch",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"147": {
"content": "gh",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"148": {
"content": "kh",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"149": {
"content": "ng",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"150": {
"content": "ngh",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"151": {
"content": "nh",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"152": {
"content": "ph",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"153": {
"content": "th",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"154": {
"content": "tr",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"155": {
"content": "gi",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"156": {
"content": "qu",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
}
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"mask_token": "<mask>",
"model_max_length": 600,
"normalize": false,
"pad_token": "<pad>",
"processor_class": "SpeechT5Processor",
"sp_model_kwargs": {},
"tokenizer_class": "SpeechT5Tokenizer",
"unk_token": "<unk>"
}