Fatou commited on
Commit
ac844d7
1 Parent(s): a496579

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 +37 -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 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 34,
3
+ "[UNK]": 33,
4
+ "a": 20,
5
+ "b": 27,
6
+ "c": 13,
7
+ "d": 32,
8
+ "e": 0,
9
+ "f": 28,
10
+ "g": 25,
11
+ "h": 22,
12
+ "i": 24,
13
+ "j": 9,
14
+ "k": 23,
15
+ "l": 26,
16
+ "m": 15,
17
+ "n": 16,
18
+ "o": 14,
19
+ "p": 7,
20
+ "q": 31,
21
+ "r": 3,
22
+ "s": 5,
23
+ "t": 6,
24
+ "u": 8,
25
+ "v": 17,
26
+ "w": 18,
27
+ "x": 21,
28
+ "y": 2,
29
+ "|": 29,
30
+ "à": 1,
31
+ "ã": 12,
32
+ "ç": 11,
33
+ "é": 30,
34
+ "ë": 4,
35
+ "ñ": 10,
36
+ "ó": 19
37
+ }