nrshoudi commited on
Commit
a3a9178
1 Parent(s): 095174b

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +12 -0
  3. vocab.json +40 -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 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "do_lower_case": false,
4
+ "eos_token": "</s>",
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "pad_token": "<PAD>",
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "replace_word_delimiter_char": " ",
9
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
10
+ "unk_token": "<UNK>",
11
+ "word_delimiter_token": "|"
12
+ }
vocab.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<PAD>": 37,
3
+ "<UNK>": 36,
4
+ "|": 0,
5
+ "آ": 1,
6
+ "أ": 2,
7
+ "إ": 3,
8
+ "ا": 4,
9
+ "ب": 5,
10
+ "ة": 6,
11
+ "ت": 7,
12
+ "ج": 8,
13
+ "ح": 9,
14
+ "خ": 10,
15
+ "د": 11,
16
+ "ذ": 12,
17
+ "ر": 13,
18
+ "س": 14,
19
+ "ش": 15,
20
+ "ص": 16,
21
+ "ض": 17,
22
+ "ط": 18,
23
+ "ظ": 19,
24
+ "ع": 20,
25
+ "غ": 21,
26
+ "ف": 22,
27
+ "ق": 23,
28
+ "ك": 24,
29
+ "ل": 25,
30
+ "م": 26,
31
+ "ن": 27,
32
+ "ه": 28,
33
+ "و": 29,
34
+ "ى": 30,
35
+ "ي": 31,
36
+ "ً": 32,
37
+ "ُ": 33,
38
+ "ِ": 34,
39
+ "ّ": 35
40
+ }