Upload tokenizer
Browse files- tokenizer_config.json +1 -3
tokenizer_config.json
CHANGED
|
@@ -34,13 +34,11 @@
|
|
| 34 |
}
|
| 35 |
},
|
| 36 |
"bos_token": "<s>",
|
| 37 |
-
"clean_up_tokenization_spaces":
|
| 38 |
"do_lower_case": false,
|
| 39 |
"eos_token": "</s>",
|
| 40 |
-
"extra_special_tokens": {},
|
| 41 |
"model_max_length": 1000000000000000019884624838656,
|
| 42 |
"pad_token": "[PAD]",
|
| 43 |
-
"processor_class": "Wav2Vec2BertProcessor",
|
| 44 |
"replace_word_delimiter_char": " ",
|
| 45 |
"target_lang": null,
|
| 46 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
|
|
|
| 34 |
}
|
| 35 |
},
|
| 36 |
"bos_token": "<s>",
|
| 37 |
+
"clean_up_tokenization_spaces": true,
|
| 38 |
"do_lower_case": false,
|
| 39 |
"eos_token": "</s>",
|
|
|
|
| 40 |
"model_max_length": 1000000000000000019884624838656,
|
| 41 |
"pad_token": "[PAD]",
|
|
|
|
| 42 |
"replace_word_delimiter_char": " ",
|
| 43 |
"target_lang": null,
|
| 44 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|