Hugging Face's logo

amineouaki
/
wav2vec2-large-xlsr-darija

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
amineouaki commited on
Commit
216a2e9
•
1 Parent(s): 7575329

amineouaki/Morocco_dialect_model_v1

Browse files
Files changed (20) hide show
  1. README.md +72 -0
  2. added_tokens.json +4 -0
  3. config.json +108 -0
  4. model.safetensors +3 -0
  5. preprocessor_config.json +10 -0
  6. runs/May16_14-06-22_66fe5fab8e9a/events.out.tfevents.1715868431.66fe5fab8e9a.20580.0 +3 -0
  7. runs/May16_14-06-22_66fe5fab8e9a/events.out.tfevents.1715872533.66fe5fab8e9a.20580.1 +3 -0
  8. runs/May16_16-11-50_66fe5fab8e9a/events.out.tfevents.1715875995.66fe5fab8e9a.55314.0 +3 -0
  9. runs/May16_16-20-00_66fe5fab8e9a/events.out.tfevents.1715876475.66fe5fab8e9a.57988.0 +3 -0
  10. runs/May16_16-20-00_66fe5fab8e9a/events.out.tfevents.1715883442.66fe5fab8e9a.57988.1 +3 -0
  11. runs/May17_20-23-43_7a206bb9369f/events.out.tfevents.1715977502.7a206bb9369f.3048.0 +3 -0
  12. runs/May20_14-07-55_fbfb0bb7579e/events.out.tfevents.1716214165.fbfb0bb7579e.1636.0 +3 -0
  13. runs/May20_17-43-07_7db01ebbf751/events.out.tfevents.1716227073.7db01ebbf751.869.0 +3 -0
  14. runs/May22_15-34-57_578db392f7f9/events.out.tfevents.1716392182.578db392f7f9.919.0 +3 -0
  15. runs/May23_08-45-22_9353e794497d/events.out.tfevents.1716454009.9353e794497d.1212.0 +3 -0
  16. runs/May23_08-45-22_9353e794497d/events.out.tfevents.1716459966.9353e794497d.1212.1 +3 -0
  17. special_tokens_map.json +6 -0
  18. tokenizer_config.json +48 -0
  19. training_args.bin +3 -0
  20. vocab.json +43 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-xls-r-300m
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: wav2vec2-large-xlsr-darija
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
+ # wav2vec2-large-xlsr-darija
17
+
18
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.5134
21
+ - Wer: 0.6540
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0003
41
+ - train_batch_size: 16
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 32
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_steps: 500
49
+ - num_epochs: 30
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
55
+ |:-------------:|:-------:|:----:|:---------------:|:------:|
56
+ | 3.0875 | 3.0928 | 300 | 3.1389 | 1.0 |
57
+ | 2.9621 | 6.1856 | 600 | 2.9607 | 1.0 |
58
+ | 2.1208 | 9.2784 | 900 | 1.8376 | 1.0093 |
59
+ | 1.0168 | 12.3711 | 1200 | 0.8073 | 0.9085 |
60
+ | 0.5872 | 15.4639 | 1500 | 0.6516 | 0.8138 |
61
+ | 0.751 | 18.5567 | 1800 | 0.5840 | 0.7466 |
62
+ | 0.3424 | 21.6495 | 2100 | 0.5712 | 0.7158 |
63
+ | 0.4591 | 24.7423 | 2400 | 0.5368 | 0.6828 |
64
+ | 0.4904 | 27.8351 | 2700 | 0.5109 | 0.6563 |
65
+
66
+
67
+ ### Framework versions
68
+
69
+ - Transformers 4.41.0
70
+ - Pytorch 2.3.0+cu121
71
+ - Datasets 2.19.1
72
+ - Tokenizers 0.19.1
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 42,
3
+ "<s>": 41
4
+ }
config.json ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-xls-r-300m",
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": 0.1,
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
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_size": 1024,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 4096,
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.1,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "wav2vec2",
70
+ "num_adapter_layers": 3,
71
+ "num_attention_heads": 16,
72
+ "num_codevector_groups": 2,
73
+ "num_codevectors_per_group": 320,
74
+ "num_conv_pos_embedding_groups": 16,
75
+ "num_conv_pos_embeddings": 128,
76
+ "num_feat_extract_layers": 7,
77
+ "num_hidden_layers": 24,
78
+ "num_negatives": 100,
79
+ "output_hidden_size": 1024,
80
+ "pad_token_id": 40,
81
+ "proj_codevector_dim": 768,
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
+ "torch_dtype": "float32",
104
+ "transformers_version": "4.41.0",
105
+ "use_weighted_layer_sum": false,
106
+ "vocab_size": 43,
107
+ "xvector_output_dim": 512
108
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8675a2ae9b08a100b730cc5fecdaf2e2c686288a5b8096a2490019a0aa6d2521
3
+ size 1261983780
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": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
runs/May16_14-06-22_66fe5fab8e9a/events.out.tfevents.1715868431.66fe5fab8e9a.20580.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67a472364867350c1d46b20aaad9a3cfe914aff6b1072042cc9602573c6e4139
3
+ size 54305
runs/May16_14-06-22_66fe5fab8e9a/events.out.tfevents.1715872533.66fe5fab8e9a.20580.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e9fa25529fce7c1c1ff7e79342abb0f217cefb4482a1cbf60a664f2738e4fa1
3
+ size 406
runs/May16_16-11-50_66fe5fab8e9a/events.out.tfevents.1715875995.66fe5fab8e9a.55314.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12204636aa5aaa13916e13c547defe20cc483d9c20e38b02babdd73de5132266
3
+ size 88
runs/May16_16-20-00_66fe5fab8e9a/events.out.tfevents.1715876475.66fe5fab8e9a.57988.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b56e64bbb8dbb97f1d5833d207643a51747c78536503abd8e780c6be00fbeb2d
3
+ size 58106
runs/May16_16-20-00_66fe5fab8e9a/events.out.tfevents.1715883442.66fe5fab8e9a.57988.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bde48c7bc77eb91825501b2ce0906dd8fcce9c81d9b434324a76f79e542d9805
3
+ size 406
runs/May17_20-23-43_7a206bb9369f/events.out.tfevents.1715977502.7a206bb9369f.3048.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec259cf931d6db28049e3b17d03eb5f3241ebc8d4bce6f2819c3b1aa3ed6dd85
3
+ size 6599
runs/May20_14-07-55_fbfb0bb7579e/events.out.tfevents.1716214165.fbfb0bb7579e.1636.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11ce93e4caf264ff794a62c4d72810a6af028e7d42355b7a18b696adf1fcdab8
3
+ size 8255
runs/May20_17-43-07_7db01ebbf751/events.out.tfevents.1716227073.7db01ebbf751.869.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:808b77f5cdb038acc83dbbd527fa0e0349b5ae93a3d7ee78f138cb19199c9ac2
3
+ size 38114
runs/May22_15-34-57_578db392f7f9/events.out.tfevents.1716392182.578db392f7f9.919.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5dd55830d54d8dd59d24b876f72b4d68db85db8df0dc5bd7e1deb6738f30810
3
+ size 6737
runs/May23_08-45-22_9353e794497d/events.out.tfevents.1716454009.9353e794497d.1212.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cb22a49daea159ce0542e4f4db30751539377cfa509b4599b71c64086e963ed
3
+ size 70892
runs/May23_08-45-22_9353e794497d/events.out.tfevents.1716459966.9353e794497d.1212.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9cb29132214a70792b3a885de4a762c1a310cbeefe424d7861995e67f59aa82
3
+ size 406
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "[PAD]",
5
+ "unk_token": "[UNK]"
6
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "39": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "40": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "41": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "42": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<s>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "do_lower_case": false,
39
+ "eos_token": "</s>",
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "[PAD]",
42
+ "processor_class": "Wav2Vec2Processor",
43
+ "replace_word_delimiter_char": " ",
44
+ "target_lang": null,
45
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
46
+ "unk_token": "[UNK]",
47
+ "word_delimiter_token": "|"
48
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de2ebe0a5f308373ef7733c7f35269b564101e7c3870c4a6fa8b9839fe8543b5
3
+ size 5176
vocab.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "[PAD]": 40,
3
+ "[UNK]": 39,
4
+ "|": 1,
5
+ "ء": 37,
6
+ "أ": 19,
7
+ "ؤ": 3,
8
+ "إ": 14,
9
+ "ئ": 34,
10
+ "ا": 21,
11
+ "ب": 5,
12
+ "ة": 2,
13
+ "ت": 16,
14
+ "ث": 22,
15
+ "ج": 9,
16
+ "ح": 10,
17
+ "خ": 29,
18
+ "د": 33,
19
+ "ذ": 20,
20
+ "ر": 17,
21
+ "ز": 13,
22
+ "س": 27,
23
+ "ش": 23,
24
+ "ص": 8,
25
+ "ض": 30,
26
+ "ط": 6,
27
+ "ظ": 25,
28
+ "ع": 15,
29
+ "غ": 32,
30
+ "ف": 28,
31
+ "ق": 38,
32
+ "ك": 35,
33
+ "ل": 24,
34
+ "م": 11,
35
+ "ن": 7,
36
+ "ه": 4,
37
+ "و": 18,
38
+ "ى": 31,
39
+ "ي": 36,
40
+ "پ": 26,
41
+ "ڤ": 0,
42
+ "ڭ": 12
43
+ }