colerobertson
commited on
Commit
•
85d2ef2
1
Parent(s):
d08ed48
Upload tokenizer
Browse files- README.md +1 -1
- added_tokens.json +2 -2
- tokenizer_config.json +3 -3
- vocab.json +35 -28
README.md
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
-
base_model: facebook/wav2vec2-base
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
6 |
datasets:
|
7 |
- minds14
|
8 |
model-index:
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
3 |
tags:
|
4 |
- generated_from_trainer
|
5 |
+
base_model: facebook/wav2vec2-base
|
6 |
datasets:
|
7 |
- minds14
|
8 |
model-index:
|
added_tokens.json
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
{
|
2 |
-
"</s>":
|
3 |
-
"<s>":
|
4 |
}
|
|
|
1 |
{
|
2 |
+
"</s>": 38,
|
3 |
+
"<s>": 37
|
4 |
}
|
tokenizer_config.json
CHANGED
@@ -8,7 +8,7 @@
|
|
8 |
"single_word": false,
|
9 |
"special": false
|
10 |
},
|
11 |
-
"
|
12 |
"content": "[UNK]",
|
13 |
"lstrip": true,
|
14 |
"normalized": false,
|
@@ -16,7 +16,7 @@
|
|
16 |
"single_word": false,
|
17 |
"special": false
|
18 |
},
|
19 |
-
"
|
20 |
"content": "<s>",
|
21 |
"lstrip": false,
|
22 |
"normalized": false,
|
@@ -24,7 +24,7 @@
|
|
24 |
"single_word": false,
|
25 |
"special": true
|
26 |
},
|
27 |
-
"
|
28 |
"content": "</s>",
|
29 |
"lstrip": false,
|
30 |
"normalized": false,
|
|
|
8 |
"single_word": false,
|
9 |
"special": false
|
10 |
},
|
11 |
+
"36": {
|
12 |
"content": "[UNK]",
|
13 |
"lstrip": true,
|
14 |
"normalized": false,
|
|
|
16 |
"single_word": false,
|
17 |
"special": false
|
18 |
},
|
19 |
+
"37": {
|
20 |
"content": "<s>",
|
21 |
"lstrip": false,
|
22 |
"normalized": false,
|
|
|
24 |
"single_word": false,
|
25 |
"special": true
|
26 |
},
|
27 |
+
"38": {
|
28 |
"content": "</s>",
|
29 |
"lstrip": false,
|
30 |
"normalized": false,
|
vocab.json
CHANGED
@@ -1,32 +1,39 @@
|
|
1 |
{
|
2 |
-
"'":
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"[PAD]": 0,
|
4 |
-
"[UNK]":
|
5 |
-
"a":
|
6 |
-
"b":
|
7 |
-
"c":
|
8 |
-
"d":
|
9 |
-
"e":
|
10 |
-
"f":
|
11 |
-
"g":
|
12 |
-
"h":
|
13 |
-
"i":
|
14 |
"j": 17,
|
15 |
-
"k":
|
16 |
-
"l":
|
17 |
-
"m":
|
18 |
-
"n":
|
19 |
-
"o":
|
20 |
-
"p":
|
21 |
-
"q":
|
22 |
-
"r":
|
23 |
-
"s":
|
24 |
-
"t":
|
25 |
-
"u":
|
26 |
-
"v":
|
27 |
-
"w":
|
28 |
-
"x":
|
29 |
-
"y":
|
30 |
-
"z":
|
31 |
-
"|":
|
|
|
32 |
}
|
|
|
1 |
{
|
2 |
+
"'": 25,
|
3 |
+
"0": 33,
|
4 |
+
"1": 32,
|
5 |
+
"2": 10,
|
6 |
+
"3": 6,
|
7 |
+
"5": 18,
|
8 |
+
"9": 28,
|
9 |
"[PAD]": 0,
|
10 |
+
"[UNK]": 36,
|
11 |
+
"a": 29,
|
12 |
+
"b": 26,
|
13 |
+
"c": 23,
|
14 |
+
"d": 21,
|
15 |
+
"e": 24,
|
16 |
+
"f": 35,
|
17 |
+
"g": 12,
|
18 |
+
"h": 2,
|
19 |
+
"i": 8,
|
20 |
"j": 17,
|
21 |
+
"k": 13,
|
22 |
+
"l": 22,
|
23 |
+
"m": 1,
|
24 |
+
"n": 3,
|
25 |
+
"o": 5,
|
26 |
+
"p": 16,
|
27 |
+
"q": 7,
|
28 |
+
"r": 15,
|
29 |
+
"s": 4,
|
30 |
+
"t": 11,
|
31 |
+
"u": 27,
|
32 |
+
"v": 20,
|
33 |
+
"w": 31,
|
34 |
+
"x": 9,
|
35 |
+
"y": 30,
|
36 |
+
"z": 19,
|
37 |
+
"|": 14,
|
38 |
+
"£": 34
|
39 |
}
|