auliaadila
commited on
Commit
•
9a64a34
1
Parent(s):
9f9b0af
Upload tokenizer
Browse files- special_tokens_map.json +0 -16
- tokenizer_config.json +0 -1
- vocab.json +25 -25
special_tokens_map.json
CHANGED
@@ -1,20 +1,4 @@
|
|
1 |
{
|
2 |
-
"additional_special_tokens": [
|
3 |
-
{
|
4 |
-
"content": "<s>",
|
5 |
-
"lstrip": false,
|
6 |
-
"normalized": true,
|
7 |
-
"rstrip": false,
|
8 |
-
"single_word": false
|
9 |
-
},
|
10 |
-
{
|
11 |
-
"content": "</s>",
|
12 |
-
"lstrip": false,
|
13 |
-
"normalized": true,
|
14 |
-
"rstrip": false,
|
15 |
-
"single_word": false
|
16 |
-
}
|
17 |
-
],
|
18 |
"bos_token": "<s>",
|
19 |
"eos_token": "</s>",
|
20 |
"pad_token": "[PAD]",
|
|
|
1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
"bos_token": "<s>",
|
3 |
"eos_token": "</s>",
|
4 |
"pad_token": "[PAD]",
|
tokenizer_config.json
CHANGED
@@ -5,7 +5,6 @@
|
|
5 |
"eos_token": "</s>",
|
6 |
"model_max_length": 1000000000000000019884624838656,
|
7 |
"pad_token": "[PAD]",
|
8 |
-
"processor_class": "Wav2Vec2Processor",
|
9 |
"replace_word_delimiter_char": " ",
|
10 |
"target_lang": null,
|
11 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
|
|
5 |
"eos_token": "</s>",
|
6 |
"model_max_length": 1000000000000000019884624838656,
|
7 |
"pad_token": "[PAD]",
|
|
|
8 |
"replace_word_delimiter_char": " ",
|
9 |
"target_lang": null,
|
10 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
vocab.json
CHANGED
@@ -1,31 +1,31 @@
|
|
1 |
{
|
2 |
"[PAD]": 28,
|
3 |
"[UNK]": 27,
|
4 |
-
"a":
|
5 |
-
"b":
|
6 |
-
"c":
|
7 |
-
"d":
|
8 |
-
"e":
|
9 |
"f": 3,
|
10 |
-
"g":
|
11 |
-
"h":
|
12 |
-
"i":
|
13 |
-
"j":
|
14 |
-
"k":
|
15 |
-
"l":
|
16 |
-
"m":
|
17 |
-
"n":
|
18 |
-
"o":
|
19 |
-
"p":
|
20 |
-
"q":
|
21 |
-
"r":
|
22 |
-
"s":
|
23 |
-
"t":
|
24 |
-
"u":
|
25 |
-
"v":
|
26 |
"w": 13,
|
27 |
-
"x":
|
28 |
-
"y":
|
29 |
-
"z":
|
30 |
-
"|":
|
31 |
}
|
|
|
1 |
{
|
2 |
"[PAD]": 28,
|
3 |
"[UNK]": 27,
|
4 |
+
"a": 21,
|
5 |
+
"b": 7,
|
6 |
+
"c": 11,
|
7 |
+
"d": 8,
|
8 |
+
"e": 22,
|
9 |
"f": 3,
|
10 |
+
"g": 25,
|
11 |
+
"h": 4,
|
12 |
+
"i": 2,
|
13 |
+
"j": 18,
|
14 |
+
"k": 17,
|
15 |
+
"l": 15,
|
16 |
+
"m": 6,
|
17 |
+
"n": 23,
|
18 |
+
"o": 16,
|
19 |
+
"p": 1,
|
20 |
+
"q": 0,
|
21 |
+
"r": 10,
|
22 |
+
"s": 14,
|
23 |
+
"t": 19,
|
24 |
+
"u": 5,
|
25 |
+
"v": 26,
|
26 |
"w": 13,
|
27 |
+
"x": 12,
|
28 |
+
"y": 24,
|
29 |
+
"z": 20,
|
30 |
+
"|": 9
|
31 |
}
|