Hugging Face's logo

dmatekenya
/
wav2vec2-large-xls-r-300m-chichewa

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
dmatekenya commited on
Commit
f25eb7d
•
1 Parent(s): 2cd687d

Upload tokenizer

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