Hugging Face's logo

dolphinnlp
/
wav2vec2vncskh

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
Metrics Community
infi commited on
Commit
222782e
·
1 Parent(s): adb48c7

Upload tokenizer

Browse files
Files changed (4) hide show
  1. added_tokens.json +6 -0
  2. special_tokens_map.json +6 -0
  3. tokenizer_config.json +47 -0
  4. vocab.json +117 -0
added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 116,
3
+ "<s>": 115,
4
+ "[PAD]": 114,
5
+ "[UNK]": 113
6
+ }
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 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "113": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "114": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "115": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "116": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<s>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "do_lower_case": false,
39
+ "eos_token": "</s>",
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "[PAD]",
42
+ "replace_word_delimiter_char": " ",
43
+ "target_lang": null,
44
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
45
+ "unk_token": "[UNK]",
46
+ "word_delimiter_token": "|"
47
+ }
vocab.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "%": 28,
3
+ ".": 113,
4
+ "0": 50,
5
+ "1": 100,
6
+ "2": 33,
7
+ "3": 106,
8
+ "4": 21,
9
+ "5": 97,
10
+ "6": 85,
11
+ "7": 40,
12
+ "8": 67,
13
+ "<": 44,
14
+ ">": 44,
15
+ "B": 43,
16
+ "C": 31,
17
+ "H": 90,
18
+ "K": 34,
19
+ "N": 19,
20
+ "P": 95,
21
+ "T": 32,
22
+ "[PAD]": 114,
23
+ "[UNK]": 113,
24
+ "a": 42,
25
+ "b": 102,
26
+ "c": 17,
27
+ "d": 8,
28
+ "e": 13,
29
+ "f": 92,
30
+ "g": 104,
31
+ "h": 29,
32
+ "i": 75,
33
+ "k": 77,
34
+ "l": 73,
35
+ "m": 1,
36
+ "n": 72,
37
+ "o": 76,
38
+ "p": 78,
39
+ "q": 66,
40
+ "r": 25,
41
+ "s": 26,
42
+ "t": 52,
43
+ "u": 10,
44
+ "v": 57,
45
+ "w": 2,
46
+ "x": 5,
47
+ "y": 81,
48
+ "z": 24,
49
+ "à": 58,
50
+ "á": 37,
51
+ "â": 18,
52
+ "ã": 65,
53
+ "è": 7,
54
+ "é": 6,
55
+ "ê": 14,
56
+ "ì": 89,
57
+ "í": 49,
58
+ "ò": 98,
59
+ "ó": 55,
60
+ "ô": 4,
61
+ "õ": 84,
62
+ "ù": 56,
63
+ "ú": 39,
64
+ "ý": 71,
65
+ "ă": 82,
66
+ "đ": 35,
67
+ "ĩ": 62,
68
+ "ũ": 15,
69
+ "ơ": 63,
70
+ "ư": 93,
71
+ "ạ": 38,
72
+ "ả": 96,
73
+ "ấ": 53,
74
+ "ầ": 70,
75
+ "ẩ": 94,
76
+ "ẫ": 83,
77
+ "ậ": 79,
78
+ "ắ": 107,
79
+ "ằ": 86,
80
+ "ẳ": 9,
81
+ "ẵ": 69,
82
+ "ặ": 74,
83
+ "ẹ": 111,
84
+ "ẻ": 112,
85
+ "ẽ": 80,
86
+ "ế": 64,
87
+ "ề": 12,
88
+ "ể": 91,
89
+ "ễ": 46,
90
+ "ệ": 22,
91
+ "ỉ": 27,
92
+ "ị": 0,
93
+ "ọ": 60,
94
+ "ỏ": 105,
95
+ "ố": 16,
96
+ "ồ": 23,
97
+ "ổ": 51,
98
+ "ỗ": 48,
99
+ "ộ": 3,
100
+ "ớ": 99,
101
+ "ờ": 47,
102
+ "ở": 88,
103
+ "ỡ": 108,
104
+ "ợ": 41,
105
+ "ụ": 54,
106
+ "ủ": 103,
107
+ "ứ": 110,
108
+ "ừ": 101,
109
+ "ử": 59,
110
+ "ữ": 36,
111
+ "ự": 109,
112
+ "ỳ": 30,
113
+ "ỷ": 87,
114
+ "ỹ": 68,
115
+ "★": 44,
116
+ "❖": 44
117
+ }