ad019el commited on
Commit
a3c57b5
1 Parent(s): 04ee444

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +13 -0
  3. vocab.json +45 -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 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "do_lower_case": false,
5
+ "eos_token": "</s>",
6
+ "model_max_length": 1000000000000000019884624838656,
7
+ "pad_token": "[PAD]",
8
+ "replace_word_delimiter_char": " ",
9
+ "target_lang": null,
10
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
11
+ "unk_token": "[UNK]",
12
+ "word_delimiter_token": "|"
13
+ }
vocab.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 42,
3
+ "[UNK]": 41,
4
+ "|": 34,
5
+ "ء": 11,
6
+ "آ": 38,
7
+ "أ": 26,
8
+ "ؤ": 13,
9
+ "إ": 20,
10
+ "ئ": 39,
11
+ "ا": 19,
12
+ "ب": 3,
13
+ "ة": 17,
14
+ "ت": 32,
15
+ "ث": 8,
16
+ "ج": 27,
17
+ "ح": 21,
18
+ "خ": 0,
19
+ "د": 9,
20
+ "ذ": 37,
21
+ "ر": 16,
22
+ "ز": 1,
23
+ "س": 14,
24
+ "ش": 36,
25
+ "ص": 40,
26
+ "ض": 12,
27
+ "ط": 22,
28
+ "ظ": 28,
29
+ "ع": 30,
30
+ "غ": 6,
31
+ "ف": 33,
32
+ "ق": 31,
33
+ "ك": 24,
34
+ "ل": 10,
35
+ "م": 5,
36
+ "ن": 35,
37
+ "ه": 15,
38
+ "و": 7,
39
+ "ى": 25,
40
+ "ي": 2,
41
+ "ٱ": 4,
42
+ "ڤ": 18,
43
+ "ک": 29,
44
+ "ی": 23
45
+ }