speecht5_finetuned_vi / tokenizer_config.json
FredDYyy's picture
Training in progress, step 1000
916c686 verified
raw
history blame contribute delete
No virus
23.7 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
},
"73": {
"content": "é",
"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
},
"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": "à",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"82": {
"content": "á",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"83": {
"content": "ã",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"84": {
"content": "ạ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"85": {
"content": "ả",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"86": {
"content": "ă",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"87": {
"content": "ắ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"88": {
"content": "ằ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"89": {
"content": "ẳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"90": {
"content": "ẵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"91": {
"content": "ặ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"92": {
"content": "â",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"93": {
"content": "ấ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"94": {
"content": "ầ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"95": {
"content": "ẩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"96": {
"content": "ẫ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"97": {
"content": "ậ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"98": {
"content": "è",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"99": {
"content": "ẹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"100": {
"content": "ẻ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"101": {
"content": "ẽ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"102": {
"content": "ề",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"103": {
"content": "ế",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"104": {
"content": "ể",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"105": {
"content": "ễ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"106": {
"content": "ệ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"107": {
"content": "đ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"108": {
"content": "ì",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"109": {
"content": "í",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"110": {
"content": "ĩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"111": {
"content": "ỉ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"112": {
"content": "ị",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"113": {
"content": "ò",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"114": {
"content": "ó",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"115": {
"content": "õ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"116": {
"content": "ọ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"117": {
"content": "ỏ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"118": {
"content": "ô",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"119": {
"content": "ố",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"120": {
"content": "ồ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"121": {
"content": "ổ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"122": {
"content": "ỗ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"123": {
"content": "ộ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"124": {
"content": "ơ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"125": {
"content": "ớ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"126": {
"content": "ờ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"127": {
"content": "ở",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"128": {
"content": "ỡ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"129": {
"content": "ợ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"130": {
"content": "ù",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"131": {
"content": "ú",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"132": {
"content": "ũ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"133": {
"content": "ụ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"134": {
"content": "ủ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"135": {
"content": "ư",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"136": {
"content": "ứ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"137": {
"content": "ừ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"138": {
"content": "ử",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"139": {
"content": "ữ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"140": {
"content": "ự",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"141": {
"content": "ỳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"142": {
"content": "ỵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"143": {
"content": "ỷ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"144": {
"content": "ỹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"145": {
"content": "ý",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"146": {
"content": "À",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"147": {
"content": "Á",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"148": {
"content": "Ã",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"149": {
"content": "Ạ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"150": {
"content": "Ả",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"151": {
"content": "Ă",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"152": {
"content": "Ắ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"153": {
"content": "Ằ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"154": {
"content": "Ẳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"155": {
"content": "Ẵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"156": {
"content": "Ặ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"157": {
"content": "Â",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"158": {
"content": "Ấ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"159": {
"content": "Ầ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"160": {
"content": "Ẩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"161": {
"content": "Ẫ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"162": {
"content": "Ậ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"163": {
"content": "È",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"164": {
"content": "É",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"165": {
"content": "Ẹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"166": {
"content": "Ẻ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"167": {
"content": "Ẽ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"168": {
"content": "Ê",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"169": {
"content": "Ề",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"170": {
"content": "Ế",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"171": {
"content": "Ể",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"172": {
"content": "Ễ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"173": {
"content": "Ệ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"174": {
"content": "Đ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"175": {
"content": "Ì",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"176": {
"content": "Í",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"177": {
"content": "Ĩ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"178": {
"content": "Ỉ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"179": {
"content": "Ị",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"180": {
"content": "Ò",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"181": {
"content": "Ó",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"182": {
"content": "Õ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"183": {
"content": "Ọ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"184": {
"content": "Ỏ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"185": {
"content": "Ô",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"186": {
"content": "Ố",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"187": {
"content": "Ồ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"188": {
"content": "Ổ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"189": {
"content": "Ỗ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"190": {
"content": "Ộ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"191": {
"content": "Ơ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"192": {
"content": "Ớ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"193": {
"content": "Ờ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"194": {
"content": "Ở",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"195": {
"content": "Ỡ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"196": {
"content": "Ợ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"197": {
"content": "Ù",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"198": {
"content": "Ú",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"199": {
"content": "Ũ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"200": {
"content": "Ụ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"201": {
"content": "Ủ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"202": {
"content": "Ư",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"203": {
"content": "Ứ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"204": {
"content": "Ừ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"205": {
"content": "Ử",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"206": {
"content": "Ữ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"207": {
"content": "Ự",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"208": {
"content": "Ỳ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"209": {
"content": "Ỵ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"210": {
"content": "Ỷ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"211": {
"content": "Ỹ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"212": {
"content": "Ý",
"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>"
}