Hugging Face's logo

dgkim0306
/
wav2vec2-large-xls-r-300m-korean-g

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
generated_from_trainer
Inference Endpoints
Model card Files Metrics Community
dgkim0306 commited on
Commit
7acc513
β€’
1 Parent(s): 2ba287d

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +10 -0
  3. vocab.json +53 -0
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 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "do_lower_case": false,
4
+ "eos_token": "</s>",
5
+ "pad_token": "[PAD]",
6
+ "replace_word_delimiter_char": " ",
7
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
8
+ "unk_token": "[UNK]",
9
+ "word_delimiter_token": "|"
10
+ }
vocab.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 50,
3
+ "[UNK]": 49,
4
+ "|": 0,
5
+ "γ„±": 1,
6
+ "γ„²": 2,
7
+ "γ„³": 3,
8
+ "γ„΄": 4,
9
+ "γ„΅": 5,
10
+ "γ„Ά": 6,
11
+ "γ„·": 7,
12
+ "γ„Έ": 8,
13
+ "γ„Ή": 9,
14
+ "γ„Ί": 10,
15
+ "γ„»": 11,
16
+ "γ„Ό": 12,
17
+ "γ…€": 13,
18
+ "ㅁ": 14,
19
+ "γ…‚": 15,
20
+ "γ…ƒ": 16,
21
+ "γ…„": 17,
22
+ "γ……": 18,
23
+ "γ…†": 19,
24
+ "γ…‡": 20,
25
+ "γ…ˆ": 21,
26
+ "γ…‰": 22,
27
+ "γ…Š": 23,
28
+ "γ…‹": 24,
29
+ "γ…Œ": 25,
30
+ "ㅍ": 26,
31
+ "γ…Ž": 27,
32
+ "ㅏ": 28,
33
+ "ㅐ": 29,
34
+ "γ…‘": 30,
35
+ "γ…’": 31,
36
+ "γ…“": 32,
37
+ "γ…”": 33,
38
+ "γ…•": 34,
39
+ "γ…–": 35,
40
+ "γ…—": 36,
41
+ "γ…˜": 37,
42
+ "γ…™": 38,
43
+ "γ…š": 39,
44
+ "γ…›": 40,
45
+ "γ…œ": 41,
46
+ "ㅝ": 42,
47
+ "γ…ž": 43,
48
+ "γ…Ÿ": 44,
49
+ "γ… ": 45,
50
+ "γ…‘": 46,
51
+ "γ…’": 47,
52
+ "γ…£": 48
53
+ }