Hugging Face's logo

abdulelahagr
/
speech-emotion-recognition-wav2vec2

Audio Classification
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
abdulelahagr commited on
Commit
affcb5b
•
1 Parent(s): 69363d2

initial commit

Browse files
Files changed (6) hide show
  1. README.md +105 -0
  2. config.json +139 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +10 -0
  5. runs/Jul05_21-31-32_dc9e802caf08/events.out.tfevents.1720215120.dc9e802caf08.782.1 +3 -0
  6. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: jonatasgrosman/wav2vec2-large-xlsr-53-english
4
+ tags:
5
+ - audio-classification
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: speech-emotion-recognition-wav2vec2
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
+ # speech-emotion-recognition-wav2vec2
18
+
19
+ This model is a fine-tuned version of [jonatasgrosman/wav2vec2-large-xlsr-53-english](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-english) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.2842
22
+ - Accuracy: 0.9045
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: 4
43
+ - eval_batch_size: 4
44
+ - seed: 42
45
+ - gradient_accumulation_steps: 2
46
+ - total_train_batch_size: 8
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: linear
49
+ - num_epochs: 1.0
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
55
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
56
+ | 2.1026 | 0.0236 | 10 | 2.0265 | 0.1592 |
57
+ | 1.9631 | 0.0472 | 20 | 2.0125 | 0.1993 |
58
+ | 1.9106 | 0.0708 | 30 | 1.8609 | 0.2417 |
59
+ | 1.715 | 0.0943 | 40 | 1.7659 | 0.3054 |
60
+ | 1.69 | 0.1179 | 50 | 1.5524 | 0.3785 |
61
+ | 1.4684 | 0.1415 | 60 | 1.4516 | 0.4057 |
62
+ | 1.3422 | 0.1651 | 70 | 1.2702 | 0.5354 |
63
+ | 1.2358 | 0.1887 | 80 | 0.9599 | 0.6899 |
64
+ | 0.9937 | 0.2123 | 90 | 0.8447 | 0.7394 |
65
+ | 0.7604 | 0.2358 | 100 | 0.8068 | 0.7453 |
66
+ | 0.7736 | 0.2594 | 110 | 0.6561 | 0.7913 |
67
+ | 0.6573 | 0.2830 | 120 | 0.6584 | 0.7830 |
68
+ | 0.5634 | 0.3066 | 130 | 0.5564 | 0.8066 |
69
+ | 0.5353 | 0.3302 | 140 | 0.5586 | 0.8184 |
70
+ | 0.3805 | 0.3538 | 150 | 0.6575 | 0.7818 |
71
+ | 0.6584 | 0.3774 | 160 | 0.4686 | 0.8538 |
72
+ | 0.4788 | 0.4009 | 170 | 0.4533 | 0.8514 |
73
+ | 0.4123 | 0.4245 | 180 | 0.5266 | 0.8432 |
74
+ | 0.4964 | 0.4481 | 190 | 0.5038 | 0.8325 |
75
+ | 0.4489 | 0.4717 | 200 | 0.5552 | 0.8208 |
76
+ | 0.4562 | 0.4953 | 210 | 0.4075 | 0.8526 |
77
+ | 0.5362 | 0.5189 | 220 | 0.4975 | 0.8184 |
78
+ | 0.3539 | 0.5425 | 230 | 0.4947 | 0.8267 |
79
+ | 0.4726 | 0.5660 | 240 | 0.4456 | 0.8514 |
80
+ | 0.3897 | 0.5896 | 250 | 0.3567 | 0.8715 |
81
+ | 0.2817 | 0.6132 | 260 | 0.3880 | 0.8644 |
82
+ | 0.3281 | 0.6368 | 270 | 0.3902 | 0.8679 |
83
+ | 0.311 | 0.6604 | 280 | 0.3243 | 0.9021 |
84
+ | 0.1768 | 0.6840 | 290 | 0.4162 | 0.8644 |
85
+ | 0.3748 | 0.7075 | 300 | 0.4482 | 0.8644 |
86
+ | 0.588 | 0.7311 | 310 | 0.3179 | 0.8950 |
87
+ | 0.402 | 0.7547 | 320 | 0.2955 | 0.9033 |
88
+ | 0.4068 | 0.7783 | 330 | 0.3212 | 0.8962 |
89
+ | 0.3622 | 0.8019 | 340 | 0.3931 | 0.8550 |
90
+ | 0.4407 | 0.8255 | 350 | 0.3467 | 0.8644 |
91
+ | 0.3474 | 0.8491 | 360 | 0.3149 | 0.8962 |
92
+ | 0.3449 | 0.8726 | 370 | 0.2829 | 0.9033 |
93
+ | 0.2673 | 0.8962 | 380 | 0.2566 | 0.9198 |
94
+ | 0.2998 | 0.9198 | 390 | 0.2614 | 0.9127 |
95
+ | 0.2721 | 0.9434 | 400 | 0.2786 | 0.9021 |
96
+ | 0.2717 | 0.9670 | 410 | 0.2891 | 0.9021 |
97
+ | 0.3277 | 0.9906 | 420 | 0.2842 | 0.9045 |
98
+
99
+
100
+ ### Framework versions
101
+
102
+ - Transformers 4.41.2
103
+ - Pytorch 2.3.0+cu121
104
+ - Datasets 2.20.0
105
+ - Tokenizers 0.19.1
config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-english",
3
+ "activation_dropout": 0.05,
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
+ "Wav2Vec2ForSpeechClassification"
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": 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": 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
+ "finetuning_task": "wav2vec2_clf",
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.05,
59
+ "hidden_size": 1024,
60
+ "id2label": {
61
+ "0": "angry",
62
+ "1": "calm",
63
+ "2": "disgust",
64
+ "3": "fear",
65
+ "4": "happy",
66
+ "5": "neutral",
67
+ "6": "sad",
68
+ "7": "surprise"
69
+ },
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "label2id": {
73
+ "angry": 0,
74
+ "calm": 1,
75
+ "disgust": 2,
76
+ "fear": 3,
77
+ "happy": 4,
78
+ "neutral": 5,
79
+ "sad": 6,
80
+ "surprise": 7
81
+ },
82
+ "layer_norm_eps": 1e-05,
83
+ "layerdrop": 0.05,
84
+ "mask_channel_length": 10,
85
+ "mask_channel_min_space": 1,
86
+ "mask_channel_other": 0.0,
87
+ "mask_channel_prob": 0.0,
88
+ "mask_channel_selection": "static",
89
+ "mask_feature_length": 10,
90
+ "mask_feature_min_masks": 0,
91
+ "mask_feature_prob": 0.0,
92
+ "mask_time_length": 10,
93
+ "mask_time_min_masks": 2,
94
+ "mask_time_min_space": 1,
95
+ "mask_time_other": 0.0,
96
+ "mask_time_prob": 0.05,
97
+ "mask_time_selection": "static",
98
+ "model_type": "wav2vec2",
99
+ "num_adapter_layers": 3,
100
+ "num_attention_heads": 16,
101
+ "num_codevector_groups": 2,
102
+ "num_codevectors_per_group": 320,
103
+ "num_conv_pos_embedding_groups": 16,
104
+ "num_conv_pos_embeddings": 128,
105
+ "num_feat_extract_layers": 7,
106
+ "num_hidden_layers": 24,
107
+ "num_negatives": 100,
108
+ "output_hidden_size": 1024,
109
+ "pad_token_id": 0,
110
+ "pooling_mode": "mean",
111
+ "problem_type": "single_label_classification",
112
+ "proj_codevector_dim": 256,
113
+ "tdnn_dilation": [
114
+ 1,
115
+ 2,
116
+ 3,
117
+ 1,
118
+ 1
119
+ ],
120
+ "tdnn_dim": [
121
+ 512,
122
+ 512,
123
+ 512,
124
+ 512,
125
+ 1500
126
+ ],
127
+ "tdnn_kernel": [
128
+ 5,
129
+ 3,
130
+ 3,
131
+ 1,
132
+ 1
133
+ ],
134
+ "torch_dtype": "float32",
135
+ "transformers_version": "4.41.2",
136
+ "use_weighted_layer_sum": false,
137
+ "vocab_size": 33,
138
+ "xvector_output_dim": 512
139
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19e988b8d51523d4d40008a242e632d6f41d567756081b99ccc1941158292cab
3
+ size 1266038904
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "processor_class": "Wav2Vec2ProcessorWithLM",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
runs/Jul05_21-31-32_dc9e802caf08/events.out.tfevents.1720215120.dc9e802caf08.782.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6aa4a8ae25244480e96f163dd6b25c4cdcbdee0f39318c2c8d5abf9618deedab
3
+ size 29577
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afdcea673538aa6e9913a7ef08f08c1be4bfd4560cfb0c96139dc953736c7089
3
+ size 5176