Hugging Face's logo

deryauysal
/
wav2vec2-large-mms-1b-turkish-colab

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
deryauysal commited on
Commit
534440a
•
1 Parent(s): 2322eca

Upload tokenizer

Browse files
Files changed (2) hide show
  1. added_tokens.json +2 -2
  2. vocab.json +82 -36
added_tokens.json CHANGED
@@ -1,4 +1,4 @@
1
  {
2
- "</s>": 38,
3
- "<s>": 37
4
  }
 
1
  {
2
+ "</s>": 84,
3
+ "<s>": 83
4
  }
vocab.json CHANGED
@@ -1,41 +1,87 @@
1
  {
2
  "tur": {
3
- "[PAD]": 36,
4
- "[UNK]": 35,
5
- "a": 1,
6
- "b": 2,
7
- "c": 3,
8
- "d": 4,
9
- "e": 5,
10
- "f": 6,
11
- "g": 7,
12
- "h": 8,
13
- "i": 9,
14
- "j": 10,
15
- "k": 11,
16
- "l": 12,
17
- "m": 13,
18
- "n": 14,
19
- "o": 15,
20
- "p": 16,
21
- "q": 17,
22
- "r": 18,
23
- "s": 19,
24
- "t": 20,
25
- "u": 21,
26
- "v": 22,
27
- "w": 23,
28
- "x": 24,
29
- "y": 25,
30
- "z": 26,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
31
  "|": 0,
32
- "ç": 27,
33
- "ë": 28,
34
- "ö": 29,
35
- "ü": 30,
36
- "ğ": 31,
37
- "ı": 32,
38
- "ş": 33,
39
- "̇": 34
 
 
 
 
 
 
 
 
 
 
 
 
40
  }
41
  }
 
1
  {
2
  "tur": {
3
+ "!": 1,
4
+ "\"": 2,
5
+ "'": 3,
6
+ "(": 4,
7
+ ")": 5,
8
+ ",": 6,
9
+ "-": 7,
10
+ ".": 8,
11
+ ":": 9,
12
+ ";": 10,
13
+ "?": 11,
14
+ "A": 12,
15
+ "B": 13,
16
+ "C": 14,
17
+ "D": 15,
18
+ "E": 16,
19
+ "F": 17,
20
+ "G": 18,
21
+ "H": 19,
22
+ "I": 20,
23
+ "J": 21,
24
+ "K": 22,
25
+ "L": 23,
26
+ "M": 24,
27
+ "N": 25,
28
+ "O": 26,
29
+ "P": 27,
30
+ "R": 28,
31
+ "S": 29,
32
+ "T": 30,
33
+ "U": 31,
34
+ "V": 32,
35
+ "W": 33,
36
+ "Y": 34,
37
+ "Z": 35,
38
+ "[PAD]": 82,
39
+ "[UNK]": 81,
40
+ "a": 36,
41
+ "b": 37,
42
+ "c": 38,
43
+ "d": 39,
44
+ "e": 40,
45
+ "f": 41,
46
+ "g": 42,
47
+ "h": 43,
48
+ "i": 44,
49
+ "j": 45,
50
+ "k": 46,
51
+ "l": 47,
52
+ "m": 48,
53
+ "n": 49,
54
+ "o": 50,
55
+ "p": 51,
56
+ "r": 52,
57
+ "s": 53,
58
+ "t": 54,
59
+ "u": 55,
60
+ "v": 56,
61
+ "w": 57,
62
+ "x": 58,
63
+ "y": 59,
64
+ "z": 60,
65
  "|": 0,
66
+ "Â": 61,
67
+ "Ç": 62,
68
+ "Ö": 63,
69
+ "Ü": 64,
70
+ "â": 65,
71
+ "ç": 66,
72
+ "î": 67,
73
+ "ö": 68,
74
+ "û": 69,
75
+ "ü": 70,
76
+ "ğ": 71,
77
+ "İ": 72,
78
+ "ı": 73,
79
+ "Ş": 74,
80
+ "ş": 75,
81
+ "‘": 76,
82
+ "’": 77,
83
+ "“": 78,
84
+ "”": 79,
85
+ "…": 80
86
  }
87
  }