Upload processor
Browse files- preprocessor_config.json +10 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +13 -0
- vocab.json +75 -0
preprocessor_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
+
"feature_size": 1,
|
5 |
+
"padding_side": "right",
|
6 |
+
"padding_value": 0.0,
|
7 |
+
"processor_class": "Wav2Vec2Processor",
|
8 |
+
"return_attention_mask": false,
|
9 |
+
"sampling_rate": 16000
|
10 |
+
}
|
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 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"do_lower_case": false,
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"max_length": 512,
|
6 |
+
"pad_token": "[PAD]",
|
7 |
+
"padding": "max_length",
|
8 |
+
"processor_class": "Wav2Vec2Processor",
|
9 |
+
"replace_word_delimiter_char": " ",
|
10 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
11 |
+
"unk_token": "[UNK]",
|
12 |
+
"word_delimiter_token": "|"
|
13 |
+
}
|
vocab.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"0": 67,
|
3 |
+
"1": 63,
|
4 |
+
"2": 35,
|
5 |
+
"3": 46,
|
6 |
+
"4": 23,
|
7 |
+
"8": 53,
|
8 |
+
"[PAD]": 72,
|
9 |
+
"[UNK]": 71,
|
10 |
+
"a": 18,
|
11 |
+
"c": 48,
|
12 |
+
"d": 68,
|
13 |
+
"e": 29,
|
14 |
+
"f": 39,
|
15 |
+
"i": 58,
|
16 |
+
"j": 61,
|
17 |
+
"m": 21,
|
18 |
+
"n": 12,
|
19 |
+
"o": 42,
|
20 |
+
"p": 6,
|
21 |
+
"r": 31,
|
22 |
+
"s": 20,
|
23 |
+
"t": 47,
|
24 |
+
"u": 28,
|
25 |
+
"v": 60,
|
26 |
+
"|": 44,
|
27 |
+
"γ±": 16,
|
28 |
+
"γ²": 8,
|
29 |
+
"γ΄": 52,
|
30 |
+
"γ΅": 24,
|
31 |
+
"γΆ": 41,
|
32 |
+
"γ·": 14,
|
33 |
+
"γΈ": 70,
|
34 |
+
"γΉ": 57,
|
35 |
+
"γΊ": 11,
|
36 |
+
"γ»": 2,
|
37 |
+
"γΌ": 3,
|
38 |
+
"γΎ": 26,
|
39 |
+
"γ
": 4,
|
40 |
+
"γ
": 1,
|
41 |
+
"γ
": 27,
|
42 |
+
"γ
": 54,
|
43 |
+
"γ
": 25,
|
44 |
+
"γ
": 36,
|
45 |
+
"γ
": 17,
|
46 |
+
"γ
": 9,
|
47 |
+
"γ
": 30,
|
48 |
+
"γ
": 66,
|
49 |
+
"γ
": 33,
|
50 |
+
"γ
": 37,
|
51 |
+
"γ
": 22,
|
52 |
+
"γ
": 19,
|
53 |
+
"γ
": 49,
|
54 |
+
"γ
": 10,
|
55 |
+
"γ
": 32,
|
56 |
+
"γ
": 56,
|
57 |
+
"γ
": 69,
|
58 |
+
"γ
": 43,
|
59 |
+
"γ
": 62,
|
60 |
+
"γ
": 55,
|
61 |
+
"γ
": 64,
|
62 |
+
"γ
": 59,
|
63 |
+
"γ
": 13,
|
64 |
+
"γ
": 51,
|
65 |
+
"γ
": 38,
|
66 |
+
"γ
": 15,
|
67 |
+
"γ
": 40,
|
68 |
+
"γ
": 50,
|
69 |
+
"γ
": 65,
|
70 |
+
"γ
": 5,
|
71 |
+
"γ
": 34,
|
72 |
+
"γ
‘": 45,
|
73 |
+
"γ
’": 0,
|
74 |
+
"γ
£": 7
|
75 |
+
}
|