aminnaghavi commited on
Commit
858c571
1 Parent(s): 3fd49c2

add tokenizer

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