| { |
| "added_tokens_decoder": { |
| "10": { |
| "content": "sh", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "15": { |
| "content": "dh", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "19": { |
| "content": "zh", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "23": { |
| "content": "rr", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "25": { |
| "content": "th", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "28": { |
| "content": "ll", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "41": { |
| "content": "nj", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "70": { |
| "content": "gj", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "93": { |
| "content": "xh", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "94": { |
| "content": "[UNK]", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "95": { |
| "content": "[PAD]", |
| "lstrip": true, |
| "normalized": false, |
| "rstrip": true, |
| "single_word": false, |
| "special": false |
| }, |
| "96": { |
| "content": "<s>", |
| "lstrip": false, |
| "normalized": false, |
| "rstrip": false, |
| "single_word": false, |
| "special": true |
| }, |
| "97": { |
| "content": "</s>", |
| "lstrip": false, |
| "normalized": false, |
| "rstrip": false, |
| "single_word": false, |
| "special": true |
| } |
| }, |
| "bos_token": "<s>", |
| "clean_up_tokenization_spaces": false, |
| "do_lower_case": false, |
| "eos_token": "</s>", |
| "extra_special_tokens": {}, |
| "model_max_length": 1000000000000000019884624838656, |
| "pad_token": "[PAD]", |
| "replace_word_delimiter_char": " ", |
| "target_lang": "aln", |
| "tokenizer_class": "Wav2Vec2CTCTokenizer", |
| "unk_token": "[UNK]", |
| "word_delimiter_token": "|" |
| } |
|
|
