monideep2255 commited on
Commit
9df2bdc
1 Parent(s): fa847b3

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +10 -0
  3. vocab.json +34 -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 +1,10 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "do_lower_case": false,
4
+ "eos_token": "</s>",
5
+ "pad_token": "[PAD]",
6
+ "replace_word_delimiter_char": " ",
7
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
8
+ "unk_token": "[UNK]",
9
+ "word_delimiter_token": "|"
10
+ }
vocab.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "'": 10,
3
+ "1": 28,
4
+ "3": 18,
5
+ "[PAD]": 31,
6
+ "[UNK]": 30,
7
+ "a": 25,
8
+ "b": 20,
9
+ "c": 26,
10
+ "d": 11,
11
+ "e": 16,
12
+ "f": 0,
13
+ "g": 8,
14
+ "h": 19,
15
+ "i": 7,
16
+ "j": 12,
17
+ "k": 21,
18
+ "l": 27,
19
+ "m": 4,
20
+ "n": 14,
21
+ "o": 2,
22
+ "p": 1,
23
+ "q": 13,
24
+ "r": 9,
25
+ "s": 24,
26
+ "t": 6,
27
+ "u": 15,
28
+ "v": 29,
29
+ "w": 3,
30
+ "x": 5,
31
+ "y": 17,
32
+ "z": 23,
33
+ "|": 22
34
+ }