Hugging Face's logo

hf-tiny-v2
/
tiny-random-Xcodec2Model

Safetensors
xcodec2
Model card Files
xet
 Community
hf-transformers-bot commited on
Commit
5c5636a
·
verified ·
1 Parent(s): 32c2060

Update tiny models for Xcodec2Model

Browse files
Files changed (3) hide show
  1. config.json +110 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +13 -0
config.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.1,
3
+ "architectures": [
4
+ "Xcodec2Model"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "audio_channels": 1,
9
+ "downsampling_ratios": [
10
+ 2,
11
+ 2,
12
+ 4
13
+ ],
14
+ "dtype": "float32",
15
+ "encoder_hidden_size": 8,
16
+ "head_dim": 8,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 32,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 4096,
21
+ "max_position_embeddings": 4096,
22
+ "model_type": "xcodec2",
23
+ "num_attention_heads": 2,
24
+ "num_hidden_layers": 2,
25
+ "num_key_value_heads": 2,
26
+ "pad_token_id": null,
27
+ "quantization_dim": 64,
28
+ "quantization_levels": [
29
+ 4,
30
+ 4,
31
+ 4,
32
+ 4
33
+ ],
34
+ "rms_norm_eps": 1e-06,
35
+ "rope_parameters": {
36
+ "rope_theta": 10000.0,
37
+ "rope_type": "default"
38
+ },
39
+ "sampling_rate": 16000,
40
+ "semantic_model_config": {
41
+ "activation_dropout": 0.0,
42
+ "adapter_act": "relu",
43
+ "adapter_kernel_size": 3,
44
+ "adapter_stride": 2,
45
+ "add_adapter": false,
46
+ "apply_spec_augment": true,
47
+ "attention_dropout": 0.0,
48
+ "bos_token_id": 1,
49
+ "classifier_proj_size": 768,
50
+ "conformer_conv_dropout": 0.1,
51
+ "conv_depthwise_kernel_size": 31,
52
+ "ctc_loss_reduction": "sum",
53
+ "ctc_zero_infinity": false,
54
+ "eos_token_id": 2,
55
+ "feat_proj_dropout": 0.0,
56
+ "feature_projection_input_dim": 160,
57
+ "final_dropout": 0.1,
58
+ "hidden_act": "swish",
59
+ "hidden_dropout": 0.0,
60
+ "hidden_size": 32,
61
+ "initializer_range": 0.02,
62
+ "intermediate_size": 64,
63
+ "layer_norm_eps": 1e-05,
64
+ "layerdrop": 0.1,
65
+ "left_max_position_embeddings": 64,
66
+ "mask_feature_length": 10,
67
+ "mask_feature_min_masks": 0,
68
+ "mask_feature_prob": 0.0,
69
+ "mask_time_length": 10,
70
+ "mask_time_min_masks": 2,
71
+ "mask_time_prob": 0.05,
72
+ "max_source_positions": 5000,
73
+ "model_type": "wav2vec2-bert",
74
+ "num_adapter_layers": 1,
75
+ "num_attention_heads": 4,
76
+ "num_hidden_layers": 17,
77
+ "output_hidden_size": 32,
78
+ "pad_token_id": 0,
79
+ "position_embeddings_type": "relative_key",
80
+ "right_max_position_embeddings": 8,
81
+ "rotary_embedding_base": 10000,
82
+ "tdnn_dilation": [
83
+ 1,
84
+ 2,
85
+ 3,
86
+ 1,
87
+ 1
88
+ ],
89
+ "tdnn_dim": [
90
+ 512,
91
+ 512,
92
+ 512,
93
+ 512,
94
+ 1500
95
+ ],
96
+ "tdnn_kernel": [
97
+ 5,
98
+ 3,
99
+ 3,
100
+ 1,
101
+ 1
102
+ ],
103
+ "use_intermediate_ffn_before_adapter": false,
104
+ "use_weighted_layer_sum": false,
105
+ "vocab_size": null,
106
+ "xvector_output_dim": 512
107
+ },
108
+ "tie_word_embeddings": false,
109
+ "transformers_version": "5.16.0.dev0"
110
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67fb7a9bcc18c2bbd4cba23ddcadcc1636a594010cfdac3a36d6ac3dd01008c5
3
+ size 3873592
preprocessor_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "feature_extractor_type": "Xcodec2FeatureExtractor",
3
+ "feature_size": 80,
4
+ "frame_length": 400,
5
+ "frame_shift": 160,
6
+ "hop_length": 320,
7
+ "num_mel_bins": 80,
8
+ "padding_side": "right",
9
+ "padding_value": 1,
10
+ "return_attention_mask": true,
11
+ "sampling_rate": 16000,
12
+ "stride": 2
13
+ }