rgomez-itg commited on
Commit
af321e6
1 Parent(s): 9f44a89

wav2vec2-large-xlsr-gl (tokenizer)

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +6 -0
  2. tokenizer_config.json +12 -0
  3. vocab.json +37 -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 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "do_lower_case": false,
4
+ "eos_token": "</s>",
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "pad_token": "[PAD]",
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "replace_word_delimiter_char": " ",
9
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
10
+ "unk_token": "[UNK]",
11
+ "word_delimiter_token": "|"
12
+ }
vocab.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 34,
3
+ "[UNK]": 33,
4
+ "a": 17,
5
+ "b": 29,
6
+ "c": 14,
7
+ "d": 11,
8
+ "e": 27,
9
+ "f": 13,
10
+ "g": 7,
11
+ "h": 8,
12
+ "i": 18,
13
+ "j": 15,
14
+ "k": 30,
15
+ "l": 26,
16
+ "m": 10,
17
+ "n": 0,
18
+ "o": 3,
19
+ "p": 4,
20
+ "q": 20,
21
+ "r": 32,
22
+ "s": 21,
23
+ "t": 1,
24
+ "u": 5,
25
+ "v": 22,
26
+ "w": 28,
27
+ "x": 6,
28
+ "y": 16,
29
+ "z": 12,
30
+ "|": 19,
31
+ "á": 24,
32
+ "é": 2,
33
+ "í": 23,
34
+ "ñ": 31,
35
+ "ó": 9,
36
+ "ú": 25
37
+ }