Hugging Face's logo

nrshoudi
/
wav2vec2-large-xls-r-300m-Arabic-phoneme

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
8
nrshoudi commited on
Commit
012a3bc
•
1 Parent(s): 9648360

Model save

Browse files
Files changed (2) hide show
  1. README.md +38 -38
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -1,5 +1,6 @@
1
  ---
2
  license: apache-2.0
 
3
  tags:
4
  - generated_from_trainer
5
  model-index:
@@ -14,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  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.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.0955
18
- - Per: 0.0852
19
 
20
  ## Model description
21
 
@@ -36,55 +37,54 @@ More information needed
36
  The following hyperparameters were used during training:
37
  - learning_rate: 0.0005
38
  - train_batch_size: 8
39
- - eval_batch_size: 8
40
  - seed: 42
41
  - gradient_accumulation_steps: 4
42
  - total_train_batch_size: 32
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 250
46
- - num_epochs: 30
47
- - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Per |
52
  |:-------------:|:-----:|:----:|:---------------:|:------:|
53
- | 2.5617 | 1.0 | 102 | 2.2924 | 1.0 |
54
- | 2.2639 | 2.0 | 204 | 2.3079 | 1.0 |
55
- | 2.3203 | 2.99 | 306 | 2.3406 | 1.0 |
56
- | 2.2545 | 4.0 | 409 | 2.2614 | 1.0 |
57
- | 2.1643 | 5.0 | 511 | 1.7664 | 1.0 |
58
- | 1.7206 | 6.0 | 613 | 1.7111 | 1.0 |
59
- | 1.7024 | 6.99 | 715 | 1.7146 | 1.0 |
60
- | 1.719 | 8.0 | 818 | 1.7485 | 1.0 |
61
- | 1.7268 | 9.0 | 920 | 1.7347 | 1.0 |
62
- | 1.7042 | 10.0 | 1022 | 1.7078 | 1.0 |
63
- | 1.6836 | 10.99 | 1124 | 1.6868 | 1.0 |
64
- | 1.6719 | 12.0 | 1227 | 1.6798 | 1.0 |
65
- | 1.6573 | 13.0 | 1329 | 1.6622 | 1.0 |
66
- | 1.6353 | 14.0 | 1431 | 1.6339 | 1.0 |
67
- | 1.6179 | 14.99 | 1533 | 1.5896 | 1.0 |
68
- | 1.5669 | 16.0 | 1636 | 1.5226 | 0.9793 |
69
- | 1.5276 | 17.0 | 1738 | 1.4972 | 0.9533 |
70
- | 1.4857 | 18.0 | 1840 | 1.4075 | 0.9757 |
71
- | 1.4144 | 18.99 | 1942 | 1.3158 | 0.9685 |
72
- | 1.3073 | 20.0 | 2045 | 1.1385 | 0.9692 |
73
- | 1.1651 | 21.0 | 2147 | 0.9220 | 0.9523 |
74
- | 1.0036 | 22.0 | 2249 | 0.7287 | 0.8698 |
75
- | 0.7561 | 22.99 | 2351 | 0.5178 | 0.5407 |
76
- | 0.6271 | 24.0 | 2454 | 0.3903 | 0.4185 |
77
- | 0.5033 | 25.0 | 2556 | 0.2525 | 0.2356 |
78
- | 0.4183 | 26.0 | 2658 | 0.1868 | 0.1624 |
79
- | 0.351 | 26.99 | 2760 | 0.1435 | 0.1200 |
80
- | 0.3256 | 28.0 | 2863 | 0.1154 | 0.1035 |
81
- | 0.2859 | 29.0 | 2965 | 0.0996 | 0.0849 |
82
- | 0.2771 | 29.93 | 3060 | 0.0996 | 0.0910 |
83
 
84
 
85
  ### Framework versions
86
 
87
- - Transformers 4.27.4
88
- - Pytorch 2.0.0+cu118
89
- - Datasets 2.11.0
90
  - Tokenizers 0.13.3
 
1
  ---
2
  license: apache-2.0
3
+ base_model: facebook/wav2vec2-xls-r-300m
4
  tags:
5
  - generated_from_trainer
6
  model-index:
 
15
 
16
  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.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.0365
19
+ - Per: 0.0138
20
 
21
  ## Model description
22
 
 
37
  The following hyperparameters were used during training:
38
  - learning_rate: 0.0005
39
  - train_batch_size: 8
40
+ - eval_batch_size: 6
41
  - seed: 42
42
  - gradient_accumulation_steps: 4
43
  - total_train_batch_size: 32
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
  - lr_scheduler_warmup_steps: 250
47
+ - num_epochs: 30.0
 
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Per |
52
  |:-------------:|:-----:|:----:|:---------------:|:------:|
53
+ | 5.3526 | 1.0 | 102 | 2.1537 | 1.0 |
54
+ | 2.0368 | 2.0 | 204 | 2.0887 | 1.0 |
55
+ | 2.0267 | 3.0 | 306 | 2.0675 | 1.0 |
56
+ | 1.9965 | 4.0 | 408 | 1.9867 | 0.9940 |
57
+ | 1.9372 | 5.0 | 510 | 1.8724 | 0.9825 |
58
+ | 1.7146 | 6.0 | 612 | 1.6606 | 0.9913 |
59
+ | 1.6576 | 7.0 | 714 | 1.6004 | 0.9933 |
60
+ | 1.4857 | 8.0 | 816 | 1.3504 | 0.9871 |
61
+ | 1.117 | 9.0 | 918 | 0.8836 | 0.8882 |
62
+ | 0.6447 | 10.0 | 1020 | 0.4553 | 0.4092 |
63
+ | 0.3553 | 11.0 | 1122 | 0.2485 | 0.1743 |
64
+ | 0.2193 | 12.0 | 1224 | 0.1664 | 0.0711 |
65
+ | 0.1677 | 13.0 | 1326 | 0.1392 | 0.0621 |
66
+ | 0.1317 | 14.0 | 1428 | 0.1072 | 0.0418 |
67
+ | 0.1203 | 15.0 | 1530 | 0.0975 | 0.0361 |
68
+ | 0.1079 | 16.0 | 1632 | 0.0936 | 0.0314 |
69
+ | 0.0902 | 17.0 | 1734 | 0.0762 | 0.0254 |
70
+ | 0.0835 | 18.0 | 1836 | 0.0609 | 0.0229 |
71
+ | 0.0691 | 19.0 | 1938 | 0.0592 | 0.0223 |
72
+ | 0.0665 | 20.0 | 2040 | 0.0601 | 0.0229 |
73
+ | 0.0599 | 21.0 | 2142 | 0.0469 | 0.0179 |
74
+ | 0.0545 | 22.0 | 2244 | 0.0520 | 0.0203 |
75
+ | 0.047 | 23.0 | 2346 | 0.0450 | 0.0163 |
76
+ | 0.0409 | 24.0 | 2448 | 0.0449 | 0.0181 |
77
+ | 0.0401 | 25.0 | 2550 | 0.0431 | 0.0160 |
78
+ | 0.0403 | 26.0 | 2652 | 0.0419 | 0.0164 |
79
+ | 0.0351 | 27.0 | 2754 | 0.0360 | 0.0143 |
80
+ | 0.0324 | 28.0 | 2856 | 0.0386 | 0.0145 |
81
+ | 0.0311 | 29.0 | 2958 | 0.0367 | 0.0139 |
82
+ | 0.028 | 30.0 | 3060 | 0.0365 | 0.0138 |
83
 
84
 
85
  ### Framework versions
86
 
87
+ - Transformers 4.32.1
88
+ - Pytorch 2.0.1+cu118
89
+ - Datasets 1.18.3
90
  - Tokenizers 0.13.3
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a611944a9f8ef260a61cc33019e0babd371594757b682338a5c527d8c0ad8c6
3
  size 1262053549
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0749d2b87e4c308ec4b6403c8d9b8432cef0d8fd0dab92f024ae2db0eb31588
3
  size 1262053549