Commit ·
79592f2
1
Parent(s): 645f031
Create tokenizer_config.json
Browse files- tokenizer_config.json +15 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"replace_word_delimiter_char": " ",
|
| 9 |
+
"silence_token": "<sil>",
|
| 10 |
+
"spoken_noise_token": "<spn>",
|
| 11 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 12 |
+
"unk_token": "<unk>",
|
| 13 |
+
"vocab_size": 43,
|
| 14 |
+
"word_delimiter_token": "|"
|
| 15 |
+
}
|