EzraWilliam commited on
Commit
0031855
1 Parent(s): f0be312

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