Hugging Face's logo

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

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
5
nrshoudi commited on
Commit
c717b4b
•
1 Parent(s): e44e500

Model save

Browse files
Files changed (2) hide show
  1. README.md +46 -36
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
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: 1.5458
19
- - Per: 0.2292
20
 
21
  ## Model description
22
 
@@ -36,55 +36,65 @@ More information needed
36
 
37
  The following hyperparameters were used during training:
38
  - learning_rate: 0.0005
39
- - train_batch_size: 2
40
  - eval_batch_size: 6
41
  - seed: 42
42
  - gradient_accumulation_steps: 4
43
- - total_train_batch_size: 8
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
- | 3.2235 | 1.0 | 2187 | 1.9282 | 0.6005 |
54
- | 1.5785 | 2.0 | 4374 | 1.4976 | 0.4575 |
55
- | 1.2838 | 3.0 | 6561 | 1.3438 | 0.4153 |
56
- | 1.1328 | 4.0 | 8748 | 1.3677 | 0.3842 |
57
- | 1.0331 | 5.0 | 10935 | 1.2632 | 0.3575 |
58
- | 0.9463 | 6.0 | 13122 | 1.2040 | 0.3515 |
59
- | 0.8686 | 7.0 | 15309 | 1.1669 | 0.3333 |
60
- | 0.8009 | 8.0 | 17496 | 1.1435 | 0.3233 |
61
- | 0.748 | 9.0 | 19683 | 1.1334 | 0.3122 |
62
- | 0.6908 | 10.0 | 21870 | 1.1131 | 0.2925 |
63
- | 0.6339 | 11.0 | 24057 | 1.1514 | 0.2872 |
64
- | 0.5911 | 12.0 | 26244 | 1.1188 | 0.2784 |
65
- | 0.5513 | 13.0 | 28431 | 1.1334 | 0.2790 |
66
- | 0.5127 | 14.0 | 30618 | 1.1662 | 0.2694 |
67
- | 0.4782 | 15.0 | 32805 | 1.1719 | 0.2634 |
68
- | 0.4448 | 16.0 | 34992 | 1.1635 | 0.2600 |
69
- | 0.4167 | 17.0 | 37179 | 1.2173 | 0.2561 |
70
- | 0.3913 | 18.0 | 39366 | 1.2252 | 0.2542 |
71
- | 0.3639 | 19.0 | 41553 | 1.2478 | 0.2490 |
72
- | 0.3441 | 20.0 | 43740 | 1.2977 | 0.2504 |
73
- | 0.3245 | 21.0 | 45927 | 1.3575 | 0.2477 |
74
- | 0.3048 | 22.0 | 48114 | 1.3489 | 0.2420 |
75
- | 0.2885 | 23.0 | 50301 | 1.3860 | 0.2398 |
76
- | 0.2715 | 24.0 | 52488 | 1.4116 | 0.2406 |
77
- | 0.2588 | 25.0 | 54675 | 1.4896 | 0.2374 |
78
- | 0.2454 | 26.0 | 56862 | 1.4672 | 0.2324 |
79
- | 0.2344 | 27.0 | 59049 | 1.5015 | 0.2324 |
80
- | 0.2245 | 28.0 | 61236 | 1.5205 | 0.2308 |
81
- | 0.2183 | 29.0 | 63423 | 1.5248 | 0.2302 |
82
- | 0.2089 | 30.0 | 65610 | 1.5458 | 0.2292 |
 
 
 
 
 
 
 
 
 
 
83
 
84
 
85
  ### Framework versions
86
 
87
- - Transformers 4.31.0
88
  - Pytorch 2.0.1+cu118
89
  - Datasets 1.18.3
90
  - Tokenizers 0.13.3
 
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: 0.7800
19
+ - Per: 0.1135
20
 
21
  ## Model description
22
 
 
36
 
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: 40.0
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Per |
52
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
53
+ | 4.5228 | 1.0 | 546 | 2.1723 | 0.6314 |
54
+ | 1.2389 | 2.0 | 1093 | 0.9571 | 0.2597 |
55
+ | 0.7931 | 3.0 | 1640 | 0.8440 | 0.2246 |
56
+ | 0.6438 | 4.0 | 2187 | 0.7831 | 0.2045 |
57
+ | 0.5584 | 5.0 | 2733 | 0.7660 | 0.1922 |
58
+ | 0.5062 | 6.0 | 3280 | 0.7193 | 0.1724 |
59
+ | 0.4596 | 7.0 | 3827 | 0.7373 | 0.1720 |
60
+ | 0.4227 | 8.0 | 4374 | 0.6829 | 0.1629 |
61
+ | 0.3832 | 9.0 | 4920 | 0.7181 | 0.1608 |
62
+ | 0.3617 | 10.0 | 5467 | 0.7043 | 0.1591 |
63
+ | 0.3495 | 11.0 | 6014 | 0.7295 | 0.1566 |
64
+ | 0.3282 | 12.0 | 6561 | 0.6897 | 0.1508 |
65
+ | 0.3086 | 13.0 | 7107 | 0.7353 | 0.1554 |
66
+ | 0.2911 | 14.0 | 7654 | 0.7144 | 0.1477 |
67
+ | 0.2801 | 15.0 | 8201 | 0.6988 | 0.1442 |
68
+ | 0.2658 | 16.0 | 8748 | 0.7061 | 0.1475 |
69
+ | 0.252 | 17.0 | 9294 | 0.7090 | 0.1403 |
70
+ | 0.2487 | 18.0 | 9841 | 0.7032 | 0.1363 |
71
+ | 0.2363 | 19.0 | 10388 | 0.7087 | 0.1395 |
72
+ | 0.222 | 20.0 | 10935 | 0.6982 | 0.1345 |
73
+ | 0.2152 | 21.0 | 11481 | 0.6964 | 0.1361 |
74
+ | 0.2063 | 22.0 | 12028 | 0.7246 | 0.1341 |
75
+ | 0.1958 | 23.0 | 12575 | 0.7331 | 0.1347 |
76
+ | 0.1866 | 24.0 | 13122 | 0.7493 | 0.1326 |
77
+ | 0.1786 | 25.0 | 13668 | 0.7536 | 0.1381 |
78
+ | 0.1751 | 26.0 | 14215 | 0.7345 | 0.1308 |
79
+ | 0.169 | 27.0 | 14762 | 0.7274 | 0.1251 |
80
+ | 0.1616 | 28.0 | 15309 | 0.7590 | 0.1293 |
81
+ | 0.1589 | 29.0 | 15855 | 0.7330 | 0.1243 |
82
+ | 0.1495 | 30.0 | 16402 | 0.7517 | 0.1228 |
83
+ | 0.1415 | 31.0 | 16949 | 0.7454 | 0.1208 |
84
+ | 0.1376 | 32.0 | 17496 | 0.7827 | 0.1254 |
85
+ | 0.1337 | 33.0 | 18042 | 0.7523 | 0.1221 |
86
+ | 0.128 | 34.0 | 18589 | 0.7752 | 0.1208 |
87
+ | 0.1262 | 35.0 | 19136 | 0.7716 | 0.1174 |
88
+ | 0.1196 | 36.0 | 19683 | 0.7620 | 0.1164 |
89
+ | 0.1161 | 37.0 | 20229 | 0.7792 | 0.1164 |
90
+ | 0.1117 | 38.0 | 20776 | 0.7800 | 0.1140 |
91
+ | 0.1103 | 39.0 | 21323 | 0.7716 | 0.1134 |
92
+ | 0.1074 | 39.95 | 21840 | 0.7800 | 0.1135 |
93
 
94
 
95
  ### Framework versions
96
 
97
+ - Transformers 4.32.0
98
  - Pytorch 2.0.1+cu118
99
  - Datasets 1.18.3
100
  - Tokenizers 0.13.3
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:390956d529409146a7c57d65797c6b68d79627d70989774276d19d6c71734c8f
3
  size 1262098669
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2421eff14a7ae47eb80fb71e9a3d0c77b62ca2bf87f919083d9d3f5034c943d1
3
  size 1262098669