Hugging Face's logo

agent404
/
ssl-1024

Safetensors
Model card Files
xet
Community
agent404 commited on
Commit
02bf97f
·
verified ·
1 Parent(s): 6b67665

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. audio2m_speech_2nodes/README.md +10 -0
  2. audio2m_speech_2nodes/config.json +143 -0
  3. audio2m_speech_2nodes/model.safetensors +3 -0
  4. audio2m_speech_4nodes/README.md +10 -0
  5. audio2m_speech_4nodes/config.json +143 -0
  6. audio2m_speech_4nodes/model.safetensors +3 -0
audio2m_speech_2nodes/README.md ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - model_hub_mixin
4
+ - pytorch_model_hub_mixin
5
+ ---
6
+
7
+ This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
8
+ - Code: [More Information Needed]
9
+ - Paper: [More Information Needed]
10
+ - Docs: [More Information Needed]
audio2m_speech_2nodes/config.json ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label_rate": 25,
3
+ "num_codebooks": 8,
4
+ "codebook_dim": 16,
5
+ "codebook_size": 2048,
6
+ "features": [
7
+ "melspec_2048"
8
+ ],
9
+ "hop_length": 240,
10
+ "n_mels": 128,
11
+ "conv_dim": 512,
12
+ "encoder_dim": 1024,
13
+ "encoder_depth": 12,
14
+ "mask_hop": 0.4,
15
+ "mask_prob": 0.6,
16
+ "is_flash": false,
17
+ "stat": {
18
+ "melspec_2048_cnt": 14282760192,
19
+ "melspec_2048_mean": 6.768444971712967,
20
+ "melspec_2048_std": 18.417922652295623
21
+ },
22
+ "w2v2_config": {
23
+ "activation_dropout": 0.1,
24
+ "adapter_kernel_size": 3,
25
+ "adapter_stride": 2,
26
+ "add_adapter": false,
27
+ "apply_spec_augment": true,
28
+ "architectures": [
29
+ "Wav2Vec2ConformerForCTC"
30
+ ],
31
+ "attention_dropout": 0.1,
32
+ "bos_token_id": 1,
33
+ "classifier_proj_size": 256,
34
+ "codevector_dim": 768,
35
+ "conformer_conv_dropout": 0.1,
36
+ "contrastive_logits_temperature": 0.1,
37
+ "conv_bias": true,
38
+ "conv_depthwise_kernel_size": 31,
39
+ "conv_dim": [
40
+ 512,
41
+ 512,
42
+ 512,
43
+ 512,
44
+ 512,
45
+ 512,
46
+ 512
47
+ ],
48
+ "conv_kernel": [
49
+ 10,
50
+ 3,
51
+ 3,
52
+ 3,
53
+ 3,
54
+ 2,
55
+ 2
56
+ ],
57
+ "conv_stride": [
58
+ 5,
59
+ 2,
60
+ 2,
61
+ 2,
62
+ 2,
63
+ 2,
64
+ 2
65
+ ],
66
+ "ctc_loss_reduction": "sum",
67
+ "ctc_zero_infinity": false,
68
+ "diversity_loss_weight": 0.1,
69
+ "do_stable_layer_norm": true,
70
+ "eos_token_id": 2,
71
+ "feat_extract_activation": "gelu",
72
+ "feat_extract_dropout": 0.0,
73
+ "feat_extract_norm": "layer",
74
+ "feat_proj_dropout": 0.1,
75
+ "feat_quantizer_dropout": 0.0,
76
+ "final_dropout": 0.1,
77
+ "gradient_checkpointing": false,
78
+ "hidden_act": "swish",
79
+ "hidden_dropout": 0.1,
80
+ "hidden_dropout_prob": 0.1,
81
+ "hidden_size": 1024,
82
+ "initializer_range": 0.02,
83
+ "intermediate_size": 4096,
84
+ "layer_norm_eps": 1e-05,
85
+ "layerdrop": 0.0,
86
+ "mask_feature_length": 10,
87
+ "mask_feature_min_masks": 0,
88
+ "mask_feature_prob": 0.0,
89
+ "mask_time_length": 10,
90
+ "mask_time_min_masks": 2,
91
+ "mask_time_prob": 0.05,
92
+ "max_source_positions": 5000,
93
+ "model_type": "wav2vec2-conformer",
94
+ "num_adapter_layers": 3,
95
+ "num_attention_heads": 16,
96
+ "num_codevector_groups": 2,
97
+ "num_codevectors_per_group": 320,
98
+ "num_conv_pos_embedding_groups": 16,
99
+ "num_conv_pos_embeddings": 128,
100
+ "num_feat_extract_layers": 7,
101
+ "num_hidden_layers": 24,
102
+ "num_negatives": 100,
103
+ "output_hidden_size": 1024,
104
+ "pad_token_id": 0,
105
+ "position_embeddings_type": "rotary",
106
+ "proj_codevector_dim": 768,
107
+ "rotary_embedding_base": 10000,
108
+ "tdnn_dilation": [
109
+ 1,
110
+ 2,
111
+ 3,
112
+ 1,
113
+ 1
114
+ ],
115
+ "tdnn_dim": [
116
+ 512,
117
+ 512,
118
+ 512,
119
+ 512,
120
+ 1500
121
+ ],
122
+ "tdnn_kernel": [
123
+ 5,
124
+ 3,
125
+ 3,
126
+ 1,
127
+ 1
128
+ ],
129
+ "torch_dtype": "float32",
130
+ "transformers_version": "4.19.0.dev0",
131
+ "use_weighted_layer_sum": false,
132
+ "vocab_size": 32,
133
+ "xvector_output_dim": 512
134
+ },
135
+ "use_rvq_target": true,
136
+ "use_vq_target": false,
137
+ "use_encodec_target": false,
138
+ "rvq_ckpt_path": null,
139
+ "recon_loss_ratio": null,
140
+ "resume_checkpoint": null,
141
+ "rvq_n_codebooks": 8,
142
+ "rvq_multi_layer_num": 1
143
+ }
audio2m_speech_2nodes/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff6ebd6d1900206c9e3d32ef9e08563435444c41318a1112f5a972e1ea40f0b7
3
+ size 1367412264
audio2m_speech_4nodes/README.md ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - model_hub_mixin
4
+ - pytorch_model_hub_mixin
5
+ ---
6
+
7
+ This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
8
+ - Code: [More Information Needed]
9
+ - Paper: [More Information Needed]
10
+ - Docs: [More Information Needed]
audio2m_speech_4nodes/config.json ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label_rate": 25,
3
+ "num_codebooks": 8,
4
+ "codebook_dim": 16,
5
+ "codebook_size": 2048,
6
+ "features": [
7
+ "melspec_2048"
8
+ ],
9
+ "hop_length": 240,
10
+ "n_mels": 128,
11
+ "conv_dim": 512,
12
+ "encoder_dim": 1024,
13
+ "encoder_depth": 12,
14
+ "mask_hop": 0.4,
15
+ "mask_prob": 0.6,
16
+ "is_flash": false,
17
+ "stat": {
18
+ "melspec_2048_cnt": 14282760192,
19
+ "melspec_2048_mean": 6.768444971712967,
20
+ "melspec_2048_std": 18.417922652295623
21
+ },
22
+ "w2v2_config": {
23
+ "activation_dropout": 0.1,
24
+ "adapter_kernel_size": 3,
25
+ "adapter_stride": 2,
26
+ "add_adapter": false,
27
+ "apply_spec_augment": true,
28
+ "architectures": [
29
+ "Wav2Vec2ConformerForCTC"
30
+ ],
31
+ "attention_dropout": 0.1,
32
+ "bos_token_id": 1,
33
+ "classifier_proj_size": 256,
34
+ "codevector_dim": 768,
35
+ "conformer_conv_dropout": 0.1,
36
+ "contrastive_logits_temperature": 0.1,
37
+ "conv_bias": true,
38
+ "conv_depthwise_kernel_size": 31,
39
+ "conv_dim": [
40
+ 512,
41
+ 512,
42
+ 512,
43
+ 512,
44
+ 512,
45
+ 512,
46
+ 512
47
+ ],
48
+ "conv_kernel": [
49
+ 10,
50
+ 3,
51
+ 3,
52
+ 3,
53
+ 3,
54
+ 2,
55
+ 2
56
+ ],
57
+ "conv_stride": [
58
+ 5,
59
+ 2,
60
+ 2,
61
+ 2,
62
+ 2,
63
+ 2,
64
+ 2
65
+ ],
66
+ "ctc_loss_reduction": "sum",
67
+ "ctc_zero_infinity": false,
68
+ "diversity_loss_weight": 0.1,
69
+ "do_stable_layer_norm": true,
70
+ "eos_token_id": 2,
71
+ "feat_extract_activation": "gelu",
72
+ "feat_extract_dropout": 0.0,
73
+ "feat_extract_norm": "layer",
74
+ "feat_proj_dropout": 0.1,
75
+ "feat_quantizer_dropout": 0.0,
76
+ "final_dropout": 0.1,
77
+ "gradient_checkpointing": false,
78
+ "hidden_act": "swish",
79
+ "hidden_dropout": 0.1,
80
+ "hidden_dropout_prob": 0.1,
81
+ "hidden_size": 1024,
82
+ "initializer_range": 0.02,
83
+ "intermediate_size": 4096,
84
+ "layer_norm_eps": 1e-05,
85
+ "layerdrop": 0.0,
86
+ "mask_feature_length": 10,
87
+ "mask_feature_min_masks": 0,
88
+ "mask_feature_prob": 0.0,
89
+ "mask_time_length": 10,
90
+ "mask_time_min_masks": 2,
91
+ "mask_time_prob": 0.05,
92
+ "max_source_positions": 5000,
93
+ "model_type": "wav2vec2-conformer",
94
+ "num_adapter_layers": 3,
95
+ "num_attention_heads": 16,
96
+ "num_codevector_groups": 2,
97
+ "num_codevectors_per_group": 320,
98
+ "num_conv_pos_embedding_groups": 16,
99
+ "num_conv_pos_embeddings": 128,
100
+ "num_feat_extract_layers": 7,
101
+ "num_hidden_layers": 24,
102
+ "num_negatives": 100,
103
+ "output_hidden_size": 1024,
104
+ "pad_token_id": 0,
105
+ "position_embeddings_type": "rotary",
106
+ "proj_codevector_dim": 768,
107
+ "rotary_embedding_base": 10000,
108
+ "tdnn_dilation": [
109
+ 1,
110
+ 2,
111
+ 3,
112
+ 1,
113
+ 1
114
+ ],
115
+ "tdnn_dim": [
116
+ 512,
117
+ 512,
118
+ 512,
119
+ 512,
120
+ 1500
121
+ ],
122
+ "tdnn_kernel": [
123
+ 5,
124
+ 3,
125
+ 3,
126
+ 1,
127
+ 1
128
+ ],
129
+ "torch_dtype": "float32",
130
+ "transformers_version": "4.19.0.dev0",
131
+ "use_weighted_layer_sum": false,
132
+ "vocab_size": 32,
133
+ "xvector_output_dim": 512
134
+ },
135
+ "use_rvq_target": true,
136
+ "use_vq_target": false,
137
+ "use_encodec_target": false,
138
+ "rvq_ckpt_path": null,
139
+ "recon_loss_ratio": null,
140
+ "resume_checkpoint": null,
141
+ "rvq_n_codebooks": 8,
142
+ "rvq_multi_layer_num": 1
143
+ }
audio2m_speech_4nodes/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:612a886ea4af2836923ba81a8ab0d21ae037997b25112dca3b518d0e98e5461d
3
+ size 1367412264