Hugging Face's logo

Baselhany
/
wav2vec2-finetuned-arabic-basel_t1

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Inference Endpoints
Model card Files Community
Baselhany commited on
Commit
285d25a
·
verified ·
1 Parent(s): 1f96d01

Upload Wav2Vec2ForCTC

Browse files
Files changed (2) hide show
  1. config.json +11 -17
  2. model.safetensors +2 -2
config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
- "_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-arabic",
3
- "activation_dropout": 0.05,
4
  "adapter_attn_dim": null,
5
  "adapter_kernel_size": 3,
6
  "adapter_stride": 2,
@@ -43,37 +43,31 @@
43
  2
44
  ],
45
  "ctc_loss_reduction": "mean",
46
- "ctc_zero_infinity": true,
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": 0.05,
54
  "feat_quantizer_dropout": 0.0,
55
- "final_dropout": 0.0,
 
56
  "hidden_act": "gelu",
57
- "hidden_dropout": 0.05,
 
58
  "hidden_size": 1024,
59
  "initializer_range": 0.02,
60
  "intermediate_size": 4096,
61
  "layer_norm_eps": 1e-05,
62
- "layerdrop": 0.05,
63
- "mask_channel_length": 10,
64
- "mask_channel_min_space": 1,
65
- "mask_channel_other": 0.0,
66
- "mask_channel_prob": 0.0,
67
- "mask_channel_selection": "static",
68
  "mask_feature_length": 10,
69
  "mask_feature_min_masks": 0,
70
  "mask_feature_prob": 0.0,
71
  "mask_time_length": 10,
72
  "mask_time_min_masks": 2,
73
- "mask_time_min_space": 1,
74
- "mask_time_other": 0.0,
75
  "mask_time_prob": 0.05,
76
- "mask_time_selection": "static",
77
  "model_type": "wav2vec2",
78
  "num_adapter_layers": 3,
79
  "num_attention_heads": 16,
@@ -85,7 +79,7 @@
85
  "num_hidden_layers": 24,
86
  "num_negatives": 100,
87
  "output_hidden_size": 1024,
88
- "pad_token_id": 41,
89
  "proj_codevector_dim": 256,
90
  "tdnn_dilation": [
91
  1,
@@ -111,6 +105,6 @@
111
  "torch_dtype": "float32",
112
  "transformers_version": "4.45.2",
113
  "use_weighted_layer_sum": false,
114
- "vocab_size": 42,
115
  "xvector_output_dim": 512
116
  }
 
1
  {
2
+ "_name_or_path": "elgeish/wav2vec2-large-xlsr-53-arabic",
3
+ "activation_dropout": 0.1,
4
  "adapter_attn_dim": null,
5
  "adapter_kernel_size": 3,
6
  "adapter_stride": 2,
 
43
  2
44
  ],
45
  "ctc_loss_reduction": "mean",
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,
 
79
  "num_hidden_layers": 24,
80
  "num_negatives": 100,
81
  "output_hidden_size": 1024,
82
+ "pad_token_id": 38,
83
  "proj_codevector_dim": 256,
84
  "tdnn_dilation": [
85
  1,
 
105
  "torch_dtype": "float32",
106
  "transformers_version": "4.45.2",
107
  "use_weighted_layer_sum": false,
108
+ "vocab_size": 39,
109
  "xvector_output_dim": 512
110
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d0fbeb33da2c604a7f8344fbde9104a7435d77341a1255cd07150316375d6bfa
3
- size 1261979680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ad843b66aa5e21d982f8be9b59584a98aa3b04843ef3469347c1292dbcc6f4a
3
+ size 1261967380