aimate-asr / tokenizer_config.json
tyanfarm's picture
Initial upload with correct pipeline tags
2058249 verified
{
"added_tokens_decoder": {
"0": {
"content": "<pad>",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"1": {
"content": "<s>",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"2": {
"content": "</s>",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"3": {
"content": "<unk>",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"7": {
"content": "sʰ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"13": {
"content": "ə̥",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"14": {
"content": "ə̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"20": {
"content": "ĩ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"23": {
"content": "oʊ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"24": {
"content": "oʊ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"39": {
"content": "ɾ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"43": {
"content": "θʰ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"44": {
"content": "eɪ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"46": {
"content": "ŋ̍",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"50": {
"content": "aɪ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"51": {
"content": "aɪ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"54": {
"content": "əʊ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"55": {
"content": "n̩",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"58": {
"content": "dʒ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"59": {
"content": "l̩",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"60": {
"content": "ɑ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"62": {
"content": "aʊ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"65": {
"content": "ɛ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"66": {
"content": "ɔɪ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"68": {
"content": "kʰ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"71": {
"content": "tʃ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"72": {
"content": "pʰ",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"76": {
"content": "æ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"77": {
"content": "ʊ̃",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
},
"78": {
"content": "m̩",
"lstrip": true,
"normalized": false,
"rstrip": true,
"single_word": false,
"special": false
}
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": false,
"do_lower_case": false,
"eos_token": "</s>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"processor_class": "Wav2Vec2Processor",
"replace_word_delimiter_char": " ",
"target_lang": null,
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"unk_token": "<unk>",
"word_delimiter_token": "|"
}