SiRoZaRuPa commited on
Commit
4c9c047
โ€ข
1 Parent(s): d8fe6d5

Upload tokenizer

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 86,
3
+ "<s>": 85
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 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "config": null,
5
+ "do_lower_case": false,
6
+ "eos_token": "</s>",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "pad_token": "[PAD]",
9
+ "replace_word_delimiter_char": " ",
10
+ "target_lang": null,
11
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
12
+ "tokenizer_type": "wav2vec2",
13
+ "unk_token": "[UNK]",
14
+ "use_fast": false,
15
+ "word_delimiter_token": "|"
16
+ }
vocab.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 82,
3
+ "[UNK]": 81,
4
+ "[็„ก้Ÿณ]": 84,
5
+ "[้›‘้Ÿณ]": 83,
6
+ "ใ": 43,
7
+ "ใ‚": 77,
8
+ "ใƒ": 67,
9
+ "ใ„": 40,
10
+ "ใ†": 46,
11
+ "ใ‡": 72,
12
+ "ใˆ": 18,
13
+ "ใ‰": 6,
14
+ "ใŠ": 74,
15
+ "ใ‹": 57,
16
+ "ใŒ": 76,
17
+ "ใ": 62,
18
+ "ใŽ": 7,
19
+ "ใ": 33,
20
+ "ใ": 24,
21
+ "ใ‘": 71,
22
+ "ใ’": 48,
23
+ "ใ“": 64,
24
+ "ใ”": 29,
25
+ "ใ•": 59,
26
+ "ใ–": 50,
27
+ "ใ—": 70,
28
+ "ใ˜": 53,
29
+ "ใ™": 23,
30
+ "ใš": 69,
31
+ "ใ›": 14,
32
+ "ใœ": 51,
33
+ "ใ": 11,
34
+ "ใž": 56,
35
+ "ใŸ": 28,
36
+ "ใ ": 37,
37
+ "ใก": 2,
38
+ "ใข": 63,
39
+ "ใฃ": 12,
40
+ "ใค": 32,
41
+ "ใฅ": 58,
42
+ "ใฆ": 15,
43
+ "ใง": 20,
44
+ "ใจ": 21,
45
+ "ใฉ": 80,
46
+ "ใช": 79,
47
+ "ใซ": 47,
48
+ "ใฌ": 38,
49
+ "ใญ": 52,
50
+ "ใฎ": 27,
51
+ "ใฏ": 17,
52
+ "ใฐ": 73,
53
+ "ใฑ": 0,
54
+ "ใฒ": 68,
55
+ "ใณ": 65,
56
+ "ใด": 4,
57
+ "ใต": 75,
58
+ "ใถ": 30,
59
+ "ใท": 3,
60
+ "ใธ": 42,
61
+ "ใน": 10,
62
+ "ใบ": 45,
63
+ "ใป": 26,
64
+ "ใผ": 13,
65
+ "ใฝ": 1,
66
+ "ใพ": 41,
67
+ "ใฟ": 19,
68
+ "ใ‚€": 31,
69
+ "ใ‚": 34,
70
+ "ใ‚‚": 60,
71
+ "ใ‚ƒ": 39,
72
+ "ใ‚„": 55,
73
+ "ใ‚…": 5,
74
+ "ใ‚†": 44,
75
+ "ใ‚‡": 66,
76
+ "ใ‚ˆ": 49,
77
+ "ใ‚‰": 35,
78
+ "ใ‚Š": 22,
79
+ "ใ‚‹": 36,
80
+ "ใ‚Œ": 16,
81
+ "ใ‚": 54,
82
+ "ใ‚": 61,
83
+ "ใ‚’": 8,
84
+ "ใ‚“": 25,
85
+ "ใ‚”": 78,
86
+ "ใƒผ": 9
87
+ }