whisper-medium-he / tokenizer_config.json
imvladikon's picture
Training in progress, step 1000
fefe854
{
"add_bos_token": false,
"add_prefix_space": false,
"added_tokens_decoder": {
"50257": {
"content": "<|endoftext|>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"50258": {
"content": "<|startoftranscript|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50259": {
"content": "<|en|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50260": {
"content": "<|zh|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50261": {
"content": "<|de|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50262": {
"content": "<|es|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50263": {
"content": "<|ru|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50264": {
"content": "<|ko|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50265": {
"content": "<|fr|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50266": {
"content": "<|ja|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50267": {
"content": "<|pt|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50268": {
"content": "<|tr|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50269": {
"content": "<|pl|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50270": {
"content": "<|ca|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50271": {
"content": "<|nl|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50272": {
"content": "<|ar|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50273": {
"content": "<|sv|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50274": {
"content": "<|it|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50275": {
"content": "<|id|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50276": {
"content": "<|hi|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50277": {
"content": "<|fi|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50278": {
"content": "<|vi|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50279": {
"content": "<|iw|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50280": {
"content": "<|uk|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50281": {
"content": "<|el|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50282": {
"content": "<|ms|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50283": {
"content": "<|cs|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50284": {
"content": "<|ro|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50285": {
"content": "<|da|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50286": {
"content": "<|hu|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50287": {
"content": "<|ta|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50288": {
"content": "<|no|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50289": {
"content": "<|th|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50290": {
"content": "<|ur|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50291": {
"content": "<|hr|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50292": {
"content": "<|bg|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50293": {
"content": "<|lt|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50294": {
"content": "<|la|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50295": {
"content": "<|mi|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50296": {
"content": "<|ml|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50297": {
"content": "<|cy|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50298": {
"content": "<|sk|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50299": {
"content": "<|te|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50300": {
"content": "<|fa|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50301": {
"content": "<|lv|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50302": {
"content": "<|bn|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50303": {
"content": "<|sr|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50304": {
"content": "<|az|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50305": {
"content": "<|sl|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50306": {
"content": "<|kn|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50307": {
"content": "<|et|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50308": {
"content": "<|mk|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50309": {
"content": "<|br|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50310": {
"content": "<|eu|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50311": {
"content": "<|is|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50312": {
"content": "<|hy|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50313": {
"content": "<|ne|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50314": {
"content": "<|mn|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50315": {
"content": "<|bs|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50316": {
"content": "<|kk|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50317": {
"content": "<|sq|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50318": {
"content": "<|sw|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50319": {
"content": "<|gl|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50320": {
"content": "<|mr|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50321": {
"content": "<|pa|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50322": {
"content": "<|si|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50323": {
"content": "<|km|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50324": {
"content": "<|sn|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50325": {
"content": "<|yo|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50326": {
"content": "<|so|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50327": {
"content": "<|af|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50328": {
"content": "<|oc|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50329": {
"content": "<|ka|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50330": {
"content": "<|be|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50331": {
"content": "<|tg|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50332": {
"content": "<|sd|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50333": {
"content": "<|gu|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50334": {
"content": "<|am|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50335": {
"content": "<|yi|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50336": {
"content": "<|lo|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50337": {
"content": "<|uz|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50338": {
"content": "<|fo|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50339": {
"content": "<|ht|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50340": {
"content": "<|ps|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50341": {
"content": "<|tk|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50342": {
"content": "<|nn|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50343": {
"content": "<|mt|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50344": {
"content": "<|sa|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50345": {
"content": "<|lb|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50346": {
"content": "<|my|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50347": {
"content": "<|bo|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50348": {
"content": "<|tl|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50349": {
"content": "<|mg|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50350": {
"content": "<|as|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50351": {
"content": "<|tt|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50352": {
"content": "<|haw|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50353": {
"content": "<|ln|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50354": {
"content": "<|ha|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50355": {
"content": "<|ba|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50356": {
"content": "<|jw|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50357": {
"content": "<|su|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50358": {
"content": "<|translate|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50359": {
"content": "<|transcribe|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50360": {
"content": "<|startoflm|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50361": {
"content": "<|startofprev|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50362": {
"content": "<|nocaptions|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"50363": {
"content": "<|notimestamps|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"additional_special_tokens": [
"<|endoftext|>",
"<|startoftranscript|>",
"<|en|>",
"<|zh|>",
"<|de|>",
"<|es|>",
"<|ru|>",
"<|ko|>",
"<|fr|>",
"<|ja|>",
"<|pt|>",
"<|tr|>",
"<|pl|>",
"<|ca|>",
"<|nl|>",
"<|ar|>",
"<|sv|>",
"<|it|>",
"<|id|>",
"<|hi|>",
"<|fi|>",
"<|vi|>",
"<|iw|>",
"<|uk|>",
"<|el|>",
"<|ms|>",
"<|cs|>",
"<|ro|>",
"<|da|>",
"<|hu|>",
"<|ta|>",
"<|no|>",
"<|th|>",
"<|ur|>",
"<|hr|>",
"<|bg|>",
"<|lt|>",
"<|la|>",
"<|mi|>",
"<|ml|>",
"<|cy|>",
"<|sk|>",
"<|te|>",
"<|fa|>",
"<|lv|>",
"<|bn|>",
"<|sr|>",
"<|az|>",
"<|sl|>",
"<|kn|>",
"<|et|>",
"<|mk|>",
"<|br|>",
"<|eu|>",
"<|is|>",
"<|hy|>",
"<|ne|>",
"<|mn|>",
"<|bs|>",
"<|kk|>",
"<|sq|>",
"<|sw|>",
"<|gl|>",
"<|mr|>",
"<|pa|>",
"<|si|>",
"<|km|>",
"<|sn|>",
"<|yo|>",
"<|so|>",
"<|af|>",
"<|oc|>",
"<|ka|>",
"<|be|>",
"<|tg|>",
"<|sd|>",
"<|gu|>",
"<|am|>",
"<|yi|>",
"<|lo|>",
"<|uz|>",
"<|fo|>",
"<|ht|>",
"<|ps|>",
"<|tk|>",
"<|nn|>",
"<|mt|>",
"<|sa|>",
"<|lb|>",
"<|my|>",
"<|bo|>",
"<|tl|>",
"<|mg|>",
"<|as|>",
"<|tt|>",
"<|haw|>",
"<|ln|>",
"<|ha|>",
"<|ba|>",
"<|jw|>",
"<|su|>",
"<|translate|>",
"<|transcribe|>",
"<|startoflm|>",
"<|startofprev|>",
"<|nocaptions|>",
"<|notimestamps|>"
],
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"errors": "replace",
"model_max_length": 1024,
"pad_token": "<|endoftext|>",
"processor_class": "WhisperProcessor",
"return_attention_mask": false,
"tokenizer_class": "WhisperTokenizer",
"unk_token": ""
}