medaminekonyali commited on
Commit
5837fb5
1 Parent(s): 8abd94e

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +11 -0
  3. vocab.json +38 -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 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "do_lower_case": false,
4
+ "eos_token": "</s>",
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "pad_token": "[PAD]",
7
+ "replace_word_delimiter_char": " ",
8
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
9
+ "unk_token": "[UNK]",
10
+ "word_delimiter_token": "|"
11
+ }
vocab.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 35,
3
+ "[UNK]": 34,
4
+ "|": 19,
5
+ "ء": 0,
6
+ "ئ": 16,
7
+ "ا": 26,
8
+ "ب": 27,
9
+ "ة": 3,
10
+ "ت": 17,
11
+ "ث": 14,
12
+ "ج": 6,
13
+ "ح": 4,
14
+ "خ": 28,
15
+ "د": 24,
16
+ "ذ": 10,
17
+ "ر": 7,
18
+ "ز": 13,
19
+ "س": 20,
20
+ "ش": 32,
21
+ "ص": 29,
22
+ "ض": 25,
23
+ "ط": 2,
24
+ "ظ": 15,
25
+ "ع": 21,
26
+ "غ": 23,
27
+ "ف": 18,
28
+ "ق": 31,
29
+ "ك": 1,
30
+ "ل": 8,
31
+ "م": 11,
32
+ "ن": 5,
33
+ "ه": 22,
34
+ "و": 9,
35
+ "ى": 33,
36
+ "ي": 12,
37
+ "ﭬ": 30
38
+ }