xacl commited on
Commit
fc49a2b
1 Parent(s): d6e1ae8

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +13 -0
  3. vocab.json +32 -0
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "[PAD]",
5
+ "unk_token": "[UNK]"
6
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "target_lang": null,
10
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
11
+ "unk_token": "[UNK]",
12
+ "word_delimiter_token": "|"
13
+ }
vocab.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "'": 10,
3
+ "[PAD]": 29,
4
+ "[UNK]": 28,
5
+ "a": 12,
6
+ "b": 7,
7
+ "c": 9,
8
+ "d": 21,
9
+ "e": 11,
10
+ "f": 22,
11
+ "g": 15,
12
+ "h": 5,
13
+ "i": 0,
14
+ "j": 6,
15
+ "k": 8,
16
+ "l": 27,
17
+ "m": 25,
18
+ "n": 17,
19
+ "o": 20,
20
+ "p": 1,
21
+ "q": 2,
22
+ "r": 26,
23
+ "s": 23,
24
+ "t": 24,
25
+ "u": 14,
26
+ "v": 16,
27
+ "w": 13,
28
+ "x": 3,
29
+ "y": 18,
30
+ "z": 19,
31
+ "|": 4
32
+ }