BauyrjanQ commited on
Commit
d165d71
โ€ข
1 Parent(s): 75ad153

Upload tokenizer

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 76,
3
+ "<s>": 75
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": "kaz",
10
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
11
+ "unk_token": "[UNK]",
12
+ "word_delimiter_token": "|"
13
+ }
vocab.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "kaz": {
3
+ "\n": 0,
4
+ "[PAD]": 74,
5
+ "[UNK]": 73,
6
+ "f": 2,
7
+ "{": 3,
8
+ "|": 1,
9
+ "}": 4,
10
+ "ษ™": 5,
11
+ "ษต": 6,
12
+ "ะฐ": 7,
13
+ "ะฑ": 8,
14
+ "ะฒ": 9,
15
+ "ะณ": 10,
16
+ "ะด": 11,
17
+ "ะต": 12,
18
+ "ะถ": 13,
19
+ "ะท": 14,
20
+ "ะธ": 15,
21
+ "ะน": 16,
22
+ "ะบ": 17,
23
+ "ะป": 18,
24
+ "ะผ": 19,
25
+ "ะฝ": 20,
26
+ "ะพ": 21,
27
+ "ะฟ": 22,
28
+ "ั€": 23,
29
+ "ั": 24,
30
+ "ั‚": 25,
31
+ "ัƒ": 26,
32
+ "ั„": 27,
33
+ "ั…": 28,
34
+ "ั†": 29,
35
+ "ั‡": 30,
36
+ "ัˆ": 31,
37
+ "ั‰": 32,
38
+ "ัŠ": 33,
39
+ "ั‹": 34,
40
+ "ัŒ": 35,
41
+ "ั": 36,
42
+ "ัŽ": 37,
43
+ "ั": 38,
44
+ "ั‘": 39,
45
+ "ั–": 40,
46
+ "า“": 41,
47
+ "า›": 42,
48
+ "าฃ": 43,
49
+ "าฏ": 44,
50
+ "าฑ": 45,
51
+ "าป": 46,
52
+ "ำ™": 47,
53
+ "ำฉ": 48,
54
+ "๏ธ": 49,
55
+ "๏นช": 50,
56
+ "๐ŸŒ": 51,
57
+ "๐ŸŽ‰": 52,
58
+ "๐Ÿ ": 53,
59
+ "๐Ÿข": 54,
60
+ "๐Ÿญ": 55,
61
+ "๐Ÿณ": 56,
62
+ "๐Ÿ‘…": 57,
63
+ "๐Ÿ‘‘": 58,
64
+ "๐Ÿ‘ค": 59,
65
+ "๐Ÿ’ฏ": 60,
66
+ "๐Ÿ’ฒ": 61,
67
+ "๐Ÿ“†": 62,
68
+ "๐Ÿ“ˆ": 63,
69
+ "๐Ÿ“–": 64,
70
+ "๐Ÿ“": 65,
71
+ "๐Ÿ“ž": 66,
72
+ "๐Ÿ”ข": 67,
73
+ "๐Ÿ”ค": 68,
74
+ "๐Ÿ•‘": 69,
75
+ "๐Ÿ˜ท": 70,
76
+ "๐Ÿงฎ": 71,
77
+ "๐Ÿชถ": 72
78
+ }
79
+ }