ImanNalia commited on
Commit
dd669bd
1 Parent(s): e5e435a

ImanNalia/FT_Wav2Vec2_960h_No_LM

Browse files
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-base-960h
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: 960h
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # 960h
17
+
18
+ This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 1.4593
21
+ - Wer: 39.9917
22
+ - Cer: 21.1881
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 0.0001
42
+ - train_batch_size: 32
43
+ - eval_batch_size: 16
44
+ - seed: 42
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 2000
48
+ - training_steps: 5000
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
53
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
54
+ | 4.9803 | 18.52 | 1000 | 2.7949 | 100.0 | 98.8206 |
55
+ | 1.1136 | 37.04 | 2000 | 1.2272 | 41.3621 | 21.5532 |
56
+ | 0.4547 | 55.56 | 3000 | 1.3323 | 41.2514 | 21.6285 |
57
+ | 0.2591 | 74.07 | 4000 | 1.3635 | 39.3688 | 20.8403 |
58
+ | 0.2232 | 92.59 | 5000 | 1.4593 | 39.9917 | 21.1881 |
59
+
60
+
61
+ ### Framework versions
62
+
63
+ - Transformers 4.39.3
64
+ - Pytorch 1.12.1+cu116
65
+ - Datasets 2.18.0
66
+ - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base-960h",
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": false,
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": "mean",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": false,
49
+ "dropout": 0.2,
50
+ "eos_token_id": 2,
51
+ "feat_extract_activation": "gelu",
52
+ "feat_extract_dropout": 0.0,
53
+ "feat_extract_norm": "group",
54
+ "feat_proj_dropout": 0.1,
55
+ "feat_quantizer_dropout": 0.0,
56
+ "final_dropout": 0.1,
57
+ "gradient_checkpointing": false,
58
+ "hidden_act": "gelu",
59
+ "hidden_dropout": 0.1,
60
+ "hidden_dropout_prob": 0.1,
61
+ "hidden_size": 768,
62
+ "initializer_range": 0.02,
63
+ "intermediate_size": 3072,
64
+ "layer_norm_eps": 1e-05,
65
+ "layerdrop": 0.1,
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
+ "model_type": "wav2vec2",
73
+ "num_adapter_layers": 3,
74
+ "num_attention_heads": 12,
75
+ "num_codevector_groups": 2,
76
+ "num_codevectors_per_group": 320,
77
+ "num_conv_pos_embedding_groups": 16,
78
+ "num_conv_pos_embeddings": 128,
79
+ "num_feat_extract_layers": 7,
80
+ "num_hidden_layers": 12,
81
+ "num_negatives": 100,
82
+ "output_hidden_size": 768,
83
+ "pad_token_id": 29,
84
+ "proj_codevector_dim": 256,
85
+ "tdnn_dilation": [
86
+ 1,
87
+ 2,
88
+ 3,
89
+ 1,
90
+ 1
91
+ ],
92
+ "tdnn_dim": [
93
+ 512,
94
+ 512,
95
+ 512,
96
+ 512,
97
+ 1500
98
+ ],
99
+ "tdnn_kernel": [
100
+ 5,
101
+ 3,
102
+ 3,
103
+ 1,
104
+ 1
105
+ ],
106
+ "torch_dtype": "float32",
107
+ "transformers_version": "4.39.3",
108
+ "use_weighted_layer_sum": false,
109
+ "vocab_size": 32,
110
+ "xvector_output_dim": 512
111
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b1ad742cac12db0c4d813672b6a0bdd0b454dec767183f699d319d85be3c56f
3
+ size 377611072
preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
runs/Apr07_02-11-17_nttbp8sxct/events.out.tfevents.1712455930.nttbp8sxct.590.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc4115ea16f68d04c11edbfcda890f9cf5d36453c61aea4739bbeeba69bc6b2f
3
+ size 6065
runs/Apr07_02-21-15_nwli4g6zsc/events.out.tfevents.1712456522.nwli4g6zsc.699.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7405e164f286fba7847f3f068b841f85455640fdeb04a6e47089f07be474cdc3
3
+ size 6066
runs/Apr07_02-43-17_nwli4g6zsc/events.out.tfevents.1712457840.nwli4g6zsc.699.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c7207c525c73b40a165e968fe97aecc2fb1b40cb16a8c30301baf888281c70e
3
+ size 9300
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:885064157f1469d90e44ed0656bb385b75759b6b75adb1acfecb82964625b94e
3
+ size 4527