Hugging Face's logo

jcrkn
/
wav2vec2-large-xls-r-300m-breton-colab

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
jcrkn commited on
Commit
4bb84aa
•
1 Parent(s): 4e0e56f

Upload tokenizer

Browse files
Files changed (2) hide show
  1. added_tokens.json +2 -2
  2. vocab.json +40 -42
added_tokens.json CHANGED
@@ -1,4 +1,4 @@
1
  {
2
- "</s>": 44,
3
- "<s>": 43
4
  }
 
1
  {
2
+ "</s>": 42,
3
+ "<s>": 41
4
  }
vocab.json CHANGED
@@ -1,45 +1,43 @@
1
  {
2
- "'": 1,
3
- "[PAD]": 42,
4
- "[UNK]": 41,
5
- "a": 2,
6
- "b": 3,
7
- "c": 4,
8
- "d": 5,
9
- "e": 6,
10
- "f": 7,
11
- "g": 8,
12
- "h": 9,
13
- "i": 10,
14
- "j": 11,
15
- "k": 12,
16
- "l": 13,
17
- "m": 14,
18
- "n": 15,
19
- "o": 16,
20
- "p": 17,
21
- "q": 18,
22
- "r": 19,
23
- "s": 20,
24
- "t": 21,
25
- "u": 22,
26
- "v": 23,
27
- "w": 24,
28
- "x": 25,
29
- "y": 26,
30
- "z": 27,
31
  "|": 0,
32
- "à": 28,
33
- "â": 29,
34
- "ç": 30,
35
- "è": 31,
36
- "é": 32,
37
- "ê": 33,
38
- "ñ": 34,
39
- "ô": 35,
40
- "ù": 36,
41
- "û": 37,
42
- "ü": 38,
43
- "ʼ": 39,
44
- "’": 40
45
  }
 
1
  {
2
+ "[PAD]": 40,
3
+ "[UNK]": 39,
4
+ "a": 1,
5
+ "b": 2,
6
+ "c": 3,
7
+ "d": 4,
8
+ "e": 5,
9
+ "f": 6,
10
+ "g": 7,
11
+ "h": 8,
12
+ "i": 9,
13
+ "j": 10,
14
+ "k": 11,
15
+ "l": 12,
16
+ "m": 13,
17
+ "n": 14,
18
+ "o": 15,
19
+ "p": 16,
20
+ "q": 17,
21
+ "r": 18,
22
+ "s": 19,
23
+ "t": 20,
24
+ "u": 21,
25
+ "v": 22,
26
+ "w": 23,
27
+ "x": 24,
28
+ "y": 25,
29
+ "z": 26,
 
30
  "|": 0,
31
+ "à": 27,
32
+ "â": 28,
33
+ "ç": 29,
34
+ "è": 30,
35
+ "é": 31,
36
+ "ê": 32,
37
+ "ñ": 33,
38
+ "ô": 34,
39
+ "ù": 35,
40
+ "û": 36,
41
+ "ü": 37,
42
+ "’": 38
 
43
  }