Hugging Face's logo

Max200293
/
wav2vec2-efeat-300m-norwegian-colab

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Model card Files Metrics Community
Max200293 commited on
Commit
04bdb37
·
1 Parent(s): 0b816b2

Upload tokenizer

Browse files
Files changed (3) hide show
  1. added_tokens.json +2 -2
  2. tokenizer_config.json +4 -4
  3. vocab.json +33 -40
added_tokens.json CHANGED
@@ -1,4 +1,4 @@
1
  {
2
- "</s>": 42,
3
- "<s>": 41
4
  }
 
1
  {
2
+ "</s>": 35,
3
+ "<s>": 34
4
  }
tokenizer_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "added_tokens_decoder": {
3
- "39": {
4
  "content": "[UNK]",
5
  "lstrip": true,
6
  "normalized": false,
@@ -8,7 +8,7 @@
8
  "single_word": false,
9
  "special": false
10
  },
11
- "40": {
12
  "content": "[PAD]",
13
  "lstrip": true,
14
  "normalized": false,
@@ -16,7 +16,7 @@
16
  "single_word": false,
17
  "special": false
18
  },
19
- "41": {
20
  "content": "<s>",
21
  "lstrip": false,
22
  "normalized": true,
@@ -24,7 +24,7 @@
24
  "single_word": false,
25
  "special": true
26
  },
27
- "42": {
28
  "content": "</s>",
29
  "lstrip": false,
30
  "normalized": true,
 
1
  {
2
  "added_tokens_decoder": {
3
+ "32": {
4
  "content": "[UNK]",
5
  "lstrip": true,
6
  "normalized": false,
 
8
  "single_word": false,
9
  "special": false
10
  },
11
+ "33": {
12
  "content": "[PAD]",
13
  "lstrip": true,
14
  "normalized": false,
 
16
  "single_word": false,
17
  "special": false
18
  },
19
+ "34": {
20
  "content": "<s>",
21
  "lstrip": false,
22
  "normalized": true,
 
24
  "single_word": false,
25
  "special": true
26
  },
27
+ "35": {
28
  "content": "</s>",
29
  "lstrip": false,
30
  "normalized": true,
vocab.json CHANGED
@@ -1,43 +1,36 @@
1
  {
2
- "[PAD]": 40,
3
- "[UNK]": 39,
4
- "a": 1,
5
- "b": 2,
6
- "d": 3,
7
- "e": 4,
8
- "f": 5,
9
- "h": 6,
10
- "i": 7,
11
- "j": 8,
12
- "k": 9,
13
- "l": 10,
14
- "m": 11,
15
- "n": 12,
16
- "o": 13,
17
- "p": 14,
18
- "r": 15,
19
- "s": 16,
20
- "t": 17,
21
- "u": 18,
22
- "v": 19,
23
- "w": 20,
24
- "x": 21,
25
- "y": 22,
26
- "z": 23,
 
 
 
 
 
27
  "|": 0,
28
- "ø": 24,
29
- "ŋ": 25,
30
- "œ": 26,
31
- "ɑ": 27,
32
- "ɔ": 28,
33
- "ə": 29,
34
- "ɛ": 30,
35
- "ɡ": 31,
36
- "ɪ": 32,
37
- "ɹ": 33,
38
- "ʂ": 34,
39
- "ʃ": 35,
40
- "ʉ": 36,
41
- "ʊ": 37,
42
- "ː": 38
43
  }
 
1
  {
2
+ "[PAD]": 33,
3
+ "[UNK]": 32,
4
+ "\\": 1,
5
+ "_": 2,
6
+ "a": 3,
7
+ "b": 4,
8
+ "c": 5,
9
+ "d": 6,
10
+ "e": 7,
11
+ "f": 8,
12
+ "g": 9,
13
+ "h": 10,
14
+ "i": 11,
15
+ "j": 12,
16
+ "k": 13,
17
+ "l": 14,
18
+ "m": 15,
19
+ "n": 16,
20
+ "o": 17,
21
+ "p": 18,
22
+ "q": 19,
23
+ "r": 20,
24
+ "s": 21,
25
+ "t": 22,
26
+ "u": 23,
27
+ "v": 24,
28
+ "w": 25,
29
+ "x": 26,
30
+ "y": 27,
31
+ "z": 28,
32
  "|": 0,
33
+ "£": 29,
34
+ "æ": 30,
35
+ "ø": 31
 
 
 
 
 
 
 
 
 
 
 
 
36
  }