Hugging Face's logo

FatihC06
/
wav2vec2-large-xlsr-53-common_voice-turkish

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files Community
FatihC06 commited on
Commit
fcce500
·
verified ·
1 Parent(s): aa26cc3

Training in progress, step 500

Browse files
Files changed (6) hide show
  1. README.md +126 -0
  2. common_voice_tr_23660893.mp3 +0 -0
  3. config.json +117 -0
  4. model.safetensors +3 -0
  5. preprocessor_config.json +9 -0
  6. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,126 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: facebook/wav2vec2-xls-r-300m
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - wer
9
+ model-index:
10
+ - name: wav2vec2-xls-r-300m-cv7-istech
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # wav2vec2-xls-r-300m-cv7-istech
18
+
19
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.7978
22
+ - Wer: 0.7496
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: 0.0002
42
+ - train_batch_size: 8
43
+ - eval_batch_size: 8
44
+ - seed: 42
45
+ - gradient_accumulation_steps: 2
46
+ - total_train_batch_size: 16
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_steps: 20
50
+ - num_epochs: 5
51
+ - mixed_precision_training: Native AMP
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
56
+ |:-------------:|:------:|:----:|:---------------:|:------:|
57
+ | 11.1946 | 0.0803 | 20 | 6.7455 | 1.0 |
58
+ | 5.1957 | 0.1606 | 40 | 3.7715 | 1.0 |
59
+ | 3.5889 | 0.2410 | 60 | 3.4781 | 1.0 |
60
+ | 3.3656 | 0.3213 | 80 | 3.3373 | 1.0 |
61
+ | 3.5262 | 0.4016 | 100 | 3.3093 | 1.0 |
62
+ | 3.1816 | 0.4819 | 120 | 3.2444 | 1.0 |
63
+ | 3.3344 | 0.5622 | 140 | 3.2887 | 1.0 |
64
+ | 3.2522 | 0.6426 | 160 | 3.2321 | 1.0 |
65
+ | 3.1819 | 0.7229 | 180 | 3.2538 | 1.0 |
66
+ | 3.3665 | 0.8032 | 200 | 3.2018 | 1.0 |
67
+ | 3.1257 | 0.8835 | 220 | 3.1751 | 1.0 |
68
+ | 3.2982 | 0.9639 | 240 | 3.1929 | 1.0 |
69
+ | 3.2181 | 1.0442 | 260 | 3.1781 | 1.0 |
70
+ | 3.1455 | 1.1245 | 280 | 3.3508 | 1.0 |
71
+ | 3.2722 | 1.2048 | 300 | 3.2343 | 1.0 |
72
+ | 3.1217 | 1.2851 | 320 | 3.1378 | 0.9999 |
73
+ | 3.2419 | 1.3655 | 340 | 3.2108 | 0.9999 |
74
+ | 3.1825 | 1.4458 | 360 | 3.1363 | 0.9999 |
75
+ | 3.1316 | 1.5261 | 380 | 3.1685 | 0.9999 |
76
+ | 3.2373 | 1.6064 | 400 | 3.1731 | 1.0 |
77
+ | 3.0634 | 1.6867 | 420 | 3.1083 | 0.9999 |
78
+ | 3.1367 | 1.7671 | 440 | 3.1582 | 0.9999 |
79
+ | 3.0831 | 1.8474 | 460 | 3.0410 | 0.9999 |
80
+ | 2.9524 | 1.9277 | 480 | 2.9578 | 0.9999 |
81
+ | 3.1076 | 2.0080 | 500 | 2.9558 | 0.9999 |
82
+ | 2.8085 | 2.0884 | 520 | 2.8155 | 0.9999 |
83
+ | 2.7768 | 2.1687 | 540 | 2.8155 | 0.9998 |
84
+ | 2.6186 | 2.2490 | 560 | 2.5386 | 1.0 |
85
+ | 2.3452 | 2.3293 | 580 | 2.3387 | 0.9994 |
86
+ | 2.4213 | 2.4096 | 600 | 2.2745 | 0.9997 |
87
+ | 1.8149 | 2.4900 | 620 | 1.8076 | 0.9906 |
88
+ | 1.747 | 2.5703 | 640 | 1.8517 | 0.9962 |
89
+ | 1.6369 | 2.6506 | 660 | 1.4838 | 0.9662 |
90
+ | 1.3732 | 2.7309 | 680 | 1.4990 | 0.9852 |
91
+ | 1.6638 | 2.8112 | 700 | 1.3854 | 0.9383 |
92
+ | 1.1034 | 2.8916 | 720 | 1.2258 | 0.9112 |
93
+ | 1.3003 | 2.9719 | 740 | 1.3696 | 0.9174 |
94
+ | 1.2128 | 3.0522 | 760 | 1.1576 | 0.8966 |
95
+ | 0.9759 | 3.1325 | 780 | 1.1325 | 0.8838 |
96
+ | 1.2551 | 3.2129 | 800 | 1.1489 | 0.8831 |
97
+ | 0.8371 | 3.2932 | 820 | 1.0224 | 0.8460 |
98
+ | 1.0434 | 3.3735 | 840 | 1.0927 | 0.8680 |
99
+ | 1.0386 | 3.4538 | 860 | 0.9731 | 0.8383 |
100
+ | 0.879 | 3.5341 | 880 | 0.9916 | 0.8428 |
101
+ | 1.1771 | 3.6145 | 900 | 0.9857 | 0.8406 |
102
+ | 0.6948 | 3.6948 | 920 | 0.9493 | 0.8203 |
103
+ | 0.9657 | 3.7751 | 940 | 0.9960 | 0.8364 |
104
+ | 0.8991 | 3.8554 | 960 | 0.8949 | 0.8013 |
105
+ | 0.7712 | 3.9357 | 980 | 0.9039 | 0.8096 |
106
+ | 1.0684 | 4.0161 | 1000 | 0.9024 | 0.8012 |
107
+ | 0.5559 | 4.0964 | 1020 | 0.8611 | 0.7872 |
108
+ | 0.8446 | 4.1767 | 1040 | 0.9017 | 0.8021 |
109
+ | 0.7911 | 4.2570 | 1060 | 0.8596 | 0.7848 |
110
+ | 0.6178 | 4.3373 | 1080 | 0.8612 | 0.7778 |
111
+ | 0.9147 | 4.4177 | 1100 | 0.8654 | 0.7756 |
112
+ | 0.5448 | 4.4980 | 1120 | 0.8222 | 0.7649 |
113
+ | 0.7812 | 4.5783 | 1140 | 0.8337 | 0.7697 |
114
+ | 0.6784 | 4.6586 | 1160 | 0.8146 | 0.7588 |
115
+ | 0.6022 | 4.7390 | 1180 | 0.8077 | 0.7538 |
116
+ | 0.8592 | 4.8193 | 1200 | 0.8121 | 0.7621 |
117
+ | 0.4884 | 4.8996 | 1220 | 0.7982 | 0.7487 |
118
+ | 0.7429 | 4.9799 | 1240 | 0.7978 | 0.7496 |
119
+
120
+
121
+ ### Framework versions
122
+
123
+ - Transformers 4.45.2
124
+ - Pytorch 2.5.1
125
+ - Datasets 2.19.1
126
+ - Tokenizers 0.20.1
common_voice_tr_23660893.mp3 ADDED
Binary file (65.1 kB). View file
 
config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
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": 0.0,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
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": "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": 0.0,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.0,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.0,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.075,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "num_adapter_layers": 3,
80
+ "num_attention_heads": 16,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_feat_extract_layers": 7,
86
+ "num_hidden_layers": 24,
87
+ "num_negatives": 100,
88
+ "output_hidden_size": 1024,
89
+ "pad_token_id": 55,
90
+ "proj_codevector_dim": 768,
91
+ "tdnn_dilation": [
92
+ 1,
93
+ 2,
94
+ 3,
95
+ 1,
96
+ 1
97
+ ],
98
+ "tdnn_dim": [
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 512,
103
+ 1500
104
+ ],
105
+ "tdnn_kernel": [
106
+ 5,
107
+ 3,
108
+ 3,
109
+ 1,
110
+ 1
111
+ ],
112
+ "torch_dtype": "float32",
113
+ "transformers_version": "4.45.2",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 58,
116
+ "xvector_output_dim": 512
117
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b993b7a5a5c359da5a385aa71b3f50c0623808a812bc0c61d1bdfb72e929c53
3
+ size 1262045280
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": true,
8
+ "sampling_rate": 16000
9
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0efaefe4fef731a440c32308c7011dd0ad939818ebec291a481a84a91a812c92
3
+ size 5304