amit0814 commited on
Commit
04b82d5
1 Parent(s): a9beb8a

Upload tokenizer

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 74,
3
+ "<s>": 73
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "</s>",
12
+ "lstrip": false,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ }
17
+ ],
18
+ "bos_token": "<s>",
19
+ "eos_token": "</s>",
20
+ "pad_token": "[PAD]",
21
+ "unk_token": "[UNK]"
22
+ }
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 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "#": 0,
3
+ "[PAD]": 72,
4
+ "[UNK]": 71,
5
+ "a": 1,
6
+ "e": 2,
7
+ "f": 3,
8
+ "i": 4,
9
+ "l": 5,
10
+ "m": 6,
11
+ "p": 7,
12
+ "r": 8,
13
+ "u": 9,
14
+ "w": 10,
15
+ "ँ": 11,
16
+ "ं": 12,
17
+ "ः": 13,
18
+ "अ": 14,
19
+ "आ": 15,
20
+ "इ": 16,
21
+ "ई": 17,
22
+ "उ": 18,
23
+ "ऊ": 19,
24
+ "ए": 20,
25
+ "ऐ": 21,
26
+ "ऑ": 22,
27
+ "ओ": 23,
28
+ "औ": 24,
29
+ "क": 25,
30
+ "ख": 26,
31
+ "ग": 27,
32
+ "घ": 28,
33
+ "च": 29,
34
+ "छ": 30,
35
+ "ज": 31,
36
+ "झ": 32,
37
+ "ट": 33,
38
+ "ठ": 34,
39
+ "ड": 35,
40
+ "ढ": 36,
41
+ "ण": 37,
42
+ "त": 38,
43
+ "थ": 39,
44
+ "द": 40,
45
+ "ध": 41,
46
+ "न": 42,
47
+ "प": 43,
48
+ "फ": 44,
49
+ "ब": 45,
50
+ "भ": 46,
51
+ "म": 47,
52
+ "य": 48,
53
+ "र": 49,
54
+ "ल": 50,
55
+ "व": 51,
56
+ "श": 52,
57
+ "ष": 53,
58
+ "स": 54,
59
+ "ह": 55,
60
+ "़": 56,
61
+ "ा": 57,
62
+ "ि": 58,
63
+ "ी": 59,
64
+ "ु": 60,
65
+ "ू": 61,
66
+ "ृ": 62,
67
+ "े": 63,
68
+ "ै": 64,
69
+ "ॉ": 65,
70
+ "ो": 66,
71
+ "ौ": 67,
72
+ "्": 68,
73
+ "ड़": 69,
74
+ "।": 70
75
+ }