ivangtorre
commited on
Commit
•
5cc1fe2
1
Parent(s):
96b072d
Upload 6 files
Browse files- config.json +115 -0
- preprocessor_config.json +10 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +13 -0
- vocab.json +77 -0
config.json
ADDED
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_dropout": 0.0,
|
3 |
+
"adapter_kernel_size": 3,
|
4 |
+
"adapter_stride": 2,
|
5 |
+
"add_adapter": false,
|
6 |
+
"apply_spec_augment": true,
|
7 |
+
"architectures": [
|
8 |
+
"Wav2Vec2ForCTC"
|
9 |
+
],
|
10 |
+
"attention_dropout": 0.1,
|
11 |
+
"bos_token_id": 1,
|
12 |
+
"classifier_proj_size": 256,
|
13 |
+
"codevector_dim": 768,
|
14 |
+
"contrastive_logits_temperature": 0.1,
|
15 |
+
"conv_bias": true,
|
16 |
+
"conv_dim": [
|
17 |
+
512,
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512,
|
22 |
+
512,
|
23 |
+
512
|
24 |
+
],
|
25 |
+
"conv_kernel": [
|
26 |
+
10,
|
27 |
+
3,
|
28 |
+
3,
|
29 |
+
3,
|
30 |
+
3,
|
31 |
+
2,
|
32 |
+
2
|
33 |
+
],
|
34 |
+
"conv_stride": [
|
35 |
+
5,
|
36 |
+
2,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2,
|
40 |
+
2,
|
41 |
+
2
|
42 |
+
],
|
43 |
+
"ctc_loss_reduction": "sum",
|
44 |
+
"ctc_zero_infinity": false,
|
45 |
+
"diversity_loss_weight": 0.1,
|
46 |
+
"do_stable_layer_norm": true,
|
47 |
+
"eos_token_id": 2,
|
48 |
+
"feat_extract_activation": "gelu",
|
49 |
+
"feat_extract_dropout": 0.0,
|
50 |
+
"feat_extract_norm": "layer",
|
51 |
+
"feat_proj_dropout": 0.1,
|
52 |
+
"feat_quantizer_dropout": 0.0,
|
53 |
+
"final_dropout": 0.0,
|
54 |
+
"gradient_checkpointing": false,
|
55 |
+
"hidden_act": "gelu",
|
56 |
+
"hidden_dropout": 0.1,
|
57 |
+
"hidden_size": 1024,
|
58 |
+
"initializer_range": 0.02,
|
59 |
+
"intermediate_size": 4096,
|
60 |
+
"layer_norm_eps": 1e-05,
|
61 |
+
"layerdrop": 0.1,
|
62 |
+
"mask_channel_length": 10,
|
63 |
+
"mask_channel_min_space": 1,
|
64 |
+
"mask_channel_other": 0.0,
|
65 |
+
"mask_channel_prob": 0.0,
|
66 |
+
"mask_channel_selection": "static",
|
67 |
+
"mask_feature_length": 10,
|
68 |
+
"mask_feature_min_masks": 0,
|
69 |
+
"mask_feature_prob": 0.0,
|
70 |
+
"mask_time_length": 10,
|
71 |
+
"mask_time_min_masks": 2,
|
72 |
+
"mask_time_min_space": 1,
|
73 |
+
"mask_time_other": 0.0,
|
74 |
+
"mask_time_prob": 0.5,
|
75 |
+
"mask_time_selection": "static",
|
76 |
+
"model_type": "wav2vec2",
|
77 |
+
"num_adapter_layers": 3,
|
78 |
+
"num_attention_heads": 16,
|
79 |
+
"num_codevector_groups": 2,
|
80 |
+
"num_codevectors_per_group": 320,
|
81 |
+
"num_conv_pos_embedding_groups": 16,
|
82 |
+
"num_conv_pos_embeddings": 128,
|
83 |
+
"num_feat_extract_layers": 7,
|
84 |
+
"num_hidden_layers": 24,
|
85 |
+
"num_negatives": 100,
|
86 |
+
"output_hidden_size": 1024,
|
87 |
+
"pad_token_id": 0,
|
88 |
+
"proj_codevector_dim": 768,
|
89 |
+
"tdnn_dilation": [
|
90 |
+
1,
|
91 |
+
2,
|
92 |
+
3,
|
93 |
+
1,
|
94 |
+
1
|
95 |
+
],
|
96 |
+
"tdnn_dim": [
|
97 |
+
512,
|
98 |
+
512,
|
99 |
+
512,
|
100 |
+
512,
|
101 |
+
1500
|
102 |
+
],
|
103 |
+
"tdnn_kernel": [
|
104 |
+
5,
|
105 |
+
3,
|
106 |
+
3,
|
107 |
+
1,
|
108 |
+
1
|
109 |
+
],
|
110 |
+
"torch_dtype": "float32",
|
111 |
+
"transformers_version": "4.29.2",
|
112 |
+
"use_weighted_layer_sum": false,
|
113 |
+
"vocab_size": 75,
|
114 |
+
"xvector_output_dim": 512
|
115 |
+
}
|
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,
|
7 |
+
"processor_class": "Wav2Vec2Processor",
|
8 |
+
"return_attention_mask": true,
|
9 |
+
"sampling_rate": 16000
|
10 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e1fdaeb8e67af53c6489003bcfd51957b75d1111c0c59426cfcebf28f1450654
|
3 |
+
size 1262203882
|
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 |
+
"clean_up_tokenization_spaces": true,
|
4 |
+
"do_lower_case": false,
|
5 |
+
"eos_token": "</s>",
|
6 |
+
"model_max_length": 1000000000000000019884624838656,
|
7 |
+
"pad_token": "<pad>",
|
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 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<s>": 1,
|
3 |
+
"<pad>": 0,
|
4 |
+
"</s>": 2,
|
5 |
+
"<unk>": 3,
|
6 |
+
"|": 4,
|
7 |
+
"a": 5,
|
8 |
+
"e": 6,
|
9 |
+
"o": 7,
|
10 |
+
"i": 8,
|
11 |
+
"r": 9,
|
12 |
+
"u": 10,
|
13 |
+
"h": 11,
|
14 |
+
"p": 12,
|
15 |
+
"m": 13,
|
16 |
+
"k": 14,
|
17 |
+
"v": 15,
|
18 |
+
"t": 16,
|
19 |
+
"'": 17,
|
20 |
+
".": 18,
|
21 |
+
"y": 19,
|
22 |
+
"n": 20,
|
23 |
+
"é": 21,
|
24 |
+
"j": 22,
|
25 |
+
"g": 23,
|
26 |
+
"b": 24,
|
27 |
+
"á": 25,
|
28 |
+
"ñ": 26,
|
29 |
+
"d": 27,
|
30 |
+
"s": 28,
|
31 |
+
"c": 29,
|
32 |
+
"ã": 30,
|
33 |
+
"ẽ": 31,
|
34 |
+
"O": 32,
|
35 |
+
"ó": 33,
|
36 |
+
"õ": 34,
|
37 |
+
"ĩ": 35,
|
38 |
+
"ʼ": 36,
|
39 |
+
"í": 37,
|
40 |
+
"M": 38,
|
41 |
+
"H": 39,
|
42 |
+
"ú": 40,
|
43 |
+
"A": 41,
|
44 |
+
"̃": 42,
|
45 |
+
"ý": 43,
|
46 |
+
"P": 44,
|
47 |
+
"U": 45,
|
48 |
+
",": 46,
|
49 |
+
"l": 47,
|
50 |
+
"ũ": 48,
|
51 |
+
"C": 49,
|
52 |
+
"ỹ": 50,
|
53 |
+
"K": 51,
|
54 |
+
"N": 52,
|
55 |
+
"E": 53,
|
56 |
+
"I": 54,
|
57 |
+
"T": 55,
|
58 |
+
"Y": 56,
|
59 |
+
"J": 57,
|
60 |
+
"G": 58,
|
61 |
+
"R": 59,
|
62 |
+
"L": 60,
|
63 |
+
"Ñ": 61,
|
64 |
+
"V": 62,
|
65 |
+
"Á": 63,
|
66 |
+
"S": 64,
|
67 |
+
"z": 65,
|
68 |
+
"-": 66,
|
69 |
+
"?": 67,
|
70 |
+
"“": 68,
|
71 |
+
"”": 69,
|
72 |
+
"D": 70,
|
73 |
+
"B": 71,
|
74 |
+
"’": 72,
|
75 |
+
"f": 73,
|
76 |
+
"q": 74
|
77 |
+
}
|