dnikolic commited on
Commit
12aa287
1 Parent(s): dd73ca1

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 +28 -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 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 25,
3
+ "[UNK]": 24,
4
+ "a": 17,
5
+ "b": 0,
6
+ "c": 9,
7
+ "d": 16,
8
+ "e": 7,
9
+ "f": 2,
10
+ "g": 4,
11
+ "h": 13,
12
+ "i": 3,
13
+ "j": 1,
14
+ "k": 20,
15
+ "l": 21,
16
+ "m": 12,
17
+ "n": 8,
18
+ "o": 11,
19
+ "p": 15,
20
+ "r": 22,
21
+ "s": 23,
22
+ "t": 10,
23
+ "u": 19,
24
+ "v": 5,
25
+ "y": 18,
26
+ "z": 6,
27
+ "|": 14
28
+ }