Martha-987 commited on
Commit
c8f657e
1 Parent(s): 6312d2a

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