Upload 9 files
Browse files- .gitattributes +1 -0
- config.json +110 -0
- dict.ltr.txt +65 -0
- preprocessor_config.json +10 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +14 -0
- topk-500000_lm.binary +3 -0
- unigrams_500000.txt +3 -0
- vocab.json +71 -0
.gitattributes
CHANGED
@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
*_000.txt filter=lfs diff=lfs merge=lfs -text
|
37 |
*.binary filter=lfs diff=lfs merge=lfs -text
|
|
|
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
*_000.txt filter=lfs diff=lfs merge=lfs -text
|
37 |
*.binary filter=lfs diff=lfs merge=lfs -text
|
38 |
+
unigrams_500000.txt filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/wav2vec2-large-960h-lv60-self",
|
3 |
+
"activation_dropout": 0.1,
|
4 |
+
"adapter_attn_dim": null,
|
5 |
+
"adapter_kernel_size": 3,
|
6 |
+
"adapter_stride": 2,
|
7 |
+
"add_adapter": false,
|
8 |
+
"apply_spec_augment": true,
|
9 |
+
"architectures": [
|
10 |
+
"Wav2Vec2ForCTC"
|
11 |
+
],
|
12 |
+
"attention_dropout": 0.1,
|
13 |
+
"bos_token_id": 1,
|
14 |
+
"classifier_proj_size": 256,
|
15 |
+
"codevector_dim": 256,
|
16 |
+
"contrastive_logits_temperature": 0.1,
|
17 |
+
"conv_bias": true,
|
18 |
+
"conv_dim": [
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512,
|
22 |
+
512,
|
23 |
+
512,
|
24 |
+
512,
|
25 |
+
512
|
26 |
+
],
|
27 |
+
"conv_kernel": [
|
28 |
+
10,
|
29 |
+
3,
|
30 |
+
3,
|
31 |
+
3,
|
32 |
+
3,
|
33 |
+
2,
|
34 |
+
2
|
35 |
+
],
|
36 |
+
"conv_stride": [
|
37 |
+
5,
|
38 |
+
2,
|
39 |
+
2,
|
40 |
+
2,
|
41 |
+
2,
|
42 |
+
2,
|
43 |
+
2
|
44 |
+
],
|
45 |
+
"ctc_loss_reduction": "sum",
|
46 |
+
"ctc_zero_infinity": false,
|
47 |
+
"diversity_loss_weight": 0.1,
|
48 |
+
"do_stable_layer_norm": true,
|
49 |
+
"eos_token_id": 2,
|
50 |
+
"feat_extract_activation": "gelu",
|
51 |
+
"feat_extract_dropout": 0.0,
|
52 |
+
"feat_extract_norm": "layer",
|
53 |
+
"feat_proj_dropout": 0.1,
|
54 |
+
"feat_quantizer_dropout": 0.0,
|
55 |
+
"final_dropout": 0.1,
|
56 |
+
"gradient_checkpointing": false,
|
57 |
+
"hidden_act": "gelu",
|
58 |
+
"hidden_dropout": 0.1,
|
59 |
+
"hidden_dropout_prob": 0.1,
|
60 |
+
"hidden_size": 1024,
|
61 |
+
"initializer_range": 0.02,
|
62 |
+
"intermediate_size": 4096,
|
63 |
+
"layer_norm_eps": 1e-05,
|
64 |
+
"layerdrop": 0.1,
|
65 |
+
"mask_feature_length": 10,
|
66 |
+
"mask_feature_min_masks": 0,
|
67 |
+
"mask_feature_prob": 0.0,
|
68 |
+
"mask_time_length": 10,
|
69 |
+
"mask_time_min_masks": 2,
|
70 |
+
"mask_time_prob": 0.05,
|
71 |
+
"model_type": "wav2vec2",
|
72 |
+
"num_adapter_layers": 3,
|
73 |
+
"num_attention_heads": 16,
|
74 |
+
"num_codevector_groups": 2,
|
75 |
+
"num_codevectors_per_group": 320,
|
76 |
+
"num_conv_pos_embedding_groups": 16,
|
77 |
+
"num_conv_pos_embeddings": 128,
|
78 |
+
"num_feat_extract_layers": 7,
|
79 |
+
"num_hidden_layers": 24,
|
80 |
+
"num_negatives": 100,
|
81 |
+
"output_hidden_size": 1024,
|
82 |
+
"pad_token_id": 0,
|
83 |
+
"proj_codevector_dim": 256,
|
84 |
+
"tdnn_dilation": [
|
85 |
+
1,
|
86 |
+
2,
|
87 |
+
3,
|
88 |
+
1,
|
89 |
+
1
|
90 |
+
],
|
91 |
+
"tdnn_dim": [
|
92 |
+
512,
|
93 |
+
512,
|
94 |
+
512,
|
95 |
+
512,
|
96 |
+
1500
|
97 |
+
],
|
98 |
+
"tdnn_kernel": [
|
99 |
+
5,
|
100 |
+
3,
|
101 |
+
3,
|
102 |
+
1,
|
103 |
+
1
|
104 |
+
],
|
105 |
+
"torch_dtype": "float32",
|
106 |
+
"transformers_version": "4.31.0",
|
107 |
+
"use_weighted_layer_sum": false,
|
108 |
+
"vocab_size": 69,
|
109 |
+
"xvector_output_dim": 512
|
110 |
+
}
|
dict.ltr.txt
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
ଈ 0
|
2 |
+
ଜ 1
|
3 |
+
ୟ 2
|
4 |
+
କ 3
|
5 |
+
ଘ 4
|
6 |
+
଼ 5
|
7 |
+
ର 6
|
8 |
+
ଦ 7
|
9 |
+
୍ 8
|
10 |
+
େ 9
|
11 |
+
ୁ 10
|
12 |
+
ଃ 11
|
13 |
+
. 12
|
14 |
+
ୋ 13
|
15 |
+
ୱ 14
|
16 |
+
ଅ 15
|
17 |
+
ଔ 16
|
18 |
+
ଊ 17
|
19 |
+
ଏ 18
|
20 |
+
ଲ 19
|
21 |
+
ପ 20
|
22 |
+
ଟ 21
|
23 |
+
ଂ 22
|
24 |
+
ଵ 23
|
25 |
+
ୈ 24
|
26 |
+
ଧ 25
|
27 |
+
ତ 26
|
28 |
+
ଳ 27
|
29 |
+
ଷ 28
|
30 |
+
ି 29
|
31 |
+
ା 30
|
32 |
+
ମ 31
|
33 |
+
ଣ 32
|
34 |
+
ନ 33
|
35 |
+
| 34
|
36 |
+
ଁ 35
|
37 |
+
ଙ 36
|
38 |
+
ଢ଼ 37
|
39 |
+
ଖ 38
|
40 |
+
ଠ 39
|
41 |
+
ଢ 40
|
42 |
+
ୃ 41
|
43 |
+
ୌ 42
|
44 |
+
ଫ 43
|
45 |
+
ଛ 44
|
46 |
+
ଗ 45
|
47 |
+
ଝ 46
|
48 |
+
ଡ଼ 47
|
49 |
+
ଯ 48
|
50 |
+
ସ 49
|
51 |
+
ଞ 50
|
52 |
+
ଡ 51
|
53 |
+
ଋ 52
|
54 |
+
ବ 53
|
55 |
+
ଓ 54
|
56 |
+
ୀ 55
|
57 |
+
ଉ 56
|
58 |
+
ଇ 57
|
59 |
+
ଭ 58
|
60 |
+
ୂ 59
|
61 |
+
ଚ 60
|
62 |
+
ଶ 61
|
63 |
+
ହ 62
|
64 |
+
ଥ 63
|
65 |
+
ଆ 64
|
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 sha256:39ecdecd56c322ebad7199f8a9683e40e6ac5558f1e708dc71b472991c816ce6
|
3 |
+
size 1262178861
|
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 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"target_lang": null,
|
11 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
12 |
+
"unk_token": "<unk>",
|
13 |
+
"word_delimiter_token": "|"
|
14 |
+
}
|
topk-500000_lm.binary
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab98755be4d001d670838216ef76df4478db1da4a3e8b1fa4ff9fa1c8ff3882e
|
3 |
+
size 614416741
|
unigrams_500000.txt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02d012b35336276bdae392717e251663bc6aa3d03f05eaa69899ee1559a40280
|
3 |
+
size 12928063
|
vocab.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
".": 16,
|
3 |
+
"</s>": 2,
|
4 |
+
"<pad>": 0,
|
5 |
+
"<s>": 1,
|
6 |
+
"<unk>": 3,
|
7 |
+
"|": 38,
|
8 |
+
"ଁ": 39,
|
9 |
+
"ଂ": 26,
|
10 |
+
"ଃ": 15,
|
11 |
+
"ଅ": 19,
|
12 |
+
"ଆ": 68,
|
13 |
+
"ଇ": 61,
|
14 |
+
"ଈ": 4,
|
15 |
+
"ଉ": 60,
|
16 |
+
"ଊ": 21,
|
17 |
+
"ଋ": 56,
|
18 |
+
"ଏ": 22,
|
19 |
+
"ଓ": 58,
|
20 |
+
"ଔ": 20,
|
21 |
+
"କ": 7,
|
22 |
+
"ଖ": 42,
|
23 |
+
"ଗ": 49,
|
24 |
+
"ଘ": 8,
|
25 |
+
"ଙ": 40,
|
26 |
+
"ଚ": 64,
|
27 |
+
"ଛ": 48,
|
28 |
+
"ଜ": 5,
|
29 |
+
"ଝ": 50,
|
30 |
+
"ଞ": 54,
|
31 |
+
"ଟ": 25,
|
32 |
+
"ଠ": 43,
|
33 |
+
"ଡ": 55,
|
34 |
+
"ଢ": 44,
|
35 |
+
"ଣ": 36,
|
36 |
+
"ତ": 30,
|
37 |
+
"ଥ": 67,
|
38 |
+
"ଦ": 11,
|
39 |
+
"ଧ": 29,
|
40 |
+
"ନ": 37,
|
41 |
+
"ପ": 24,
|
42 |
+
"ଫ": 47,
|
43 |
+
"ବ": 57,
|
44 |
+
"ଭ": 62,
|
45 |
+
"ମ": 35,
|
46 |
+
"ଯ": 52,
|
47 |
+
"ର": 10,
|
48 |
+
"ଲ": 23,
|
49 |
+
"ଳ": 31,
|
50 |
+
"ଵ": 27,
|
51 |
+
"ଶ": 65,
|
52 |
+
"ଷ": 32,
|
53 |
+
"ସ": 53,
|
54 |
+
"ହ": 66,
|
55 |
+
"଼": 9,
|
56 |
+
"ା": 34,
|
57 |
+
"ି": 33,
|
58 |
+
"ୀ": 59,
|
59 |
+
"ୁ": 14,
|
60 |
+
"ୂ": 63,
|
61 |
+
"ୃ": 45,
|
62 |
+
"େ": 13,
|
63 |
+
"ୈ": 28,
|
64 |
+
"ୋ": 17,
|
65 |
+
"ୌ": 46,
|
66 |
+
"୍": 12,
|
67 |
+
"ଡ଼": 51,
|
68 |
+
"ଢ଼": 41,
|
69 |
+
"ୟ": 6,
|
70 |
+
"ୱ": 18
|
71 |
+
}
|