Hugging Face's logo

csikasote
/
mms-1b-toigen-balanced-model

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
toigen
mms
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
csikasote commited on
Commit
89b17c6
·
verified ·
1 Parent(s): 8ab5141

Model save

Browse files
Files changed (2) hide show
  1. README.md +30 -34
  2. model.safetensors +1 -1
README.md CHANGED
@@ -3,9 +3,6 @@ library_name: transformers
3
  license: cc-by-nc-4.0
4
  base_model: facebook/mms-1b-all
5
  tags:
6
- - automatic-speech-recognition
7
- - toigen
8
- - mms
9
  - generated_from_trainer
10
  metrics:
11
  - wer
@@ -19,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  # mms-1b-toigen-balanced-model
21
 
22
- This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the TOIGEN - TOI dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.3740
25
- - Wer: 0.3990
26
 
27
  ## Model description
28
 
@@ -45,43 +42,42 @@ The following hyperparameters were used during training:
45
  - train_batch_size: 4
46
  - eval_batch_size: 4
47
  - seed: 42
 
 
48
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
  - lr_scheduler_type: linear
50
  - lr_scheduler_warmup_steps: 100
51
- - num_epochs: 30.0
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Wer |
57
  |:-------------:|:-------:|:----:|:---------------:|:------:|
58
- | 7.7726 | 0.4464 | 100 | 3.8109 | 0.9938 |
59
- | 2.5726 | 0.8929 | 200 | 0.8106 | 0.6167 |
60
- | 0.7986 | 1.3393 | 300 | 0.5409 | 0.5258 |
61
- | 0.6324 | 1.7857 | 400 | 0.5256 | 0.5054 |
62
- | 0.603 | 2.2321 | 500 | 0.4854 | 0.4832 |
63
- | 0.59 | 2.6786 | 600 | 0.4733 | 0.4846 |
64
- | 0.5489 | 3.125 | 700 | 0.4440 | 0.4657 |
65
- | 0.5173 | 3.5714 | 800 | 0.4322 | 0.4576 |
66
- | 0.5315 | 4.0179 | 900 | 0.4286 | 0.4453 |
67
- | 0.4912 | 4.4643 | 1000 | 0.4254 | 0.4458 |
68
- | 0.4728 | 4.9107 | 1100 | 0.4346 | 0.4430 |
69
- | 0.4989 | 5.3571 | 1200 | 0.4050 | 0.4292 |
70
- | 0.4661 | 5.8036 | 1300 | 0.4019 | 0.4255 |
71
- | 0.4755 | 6.25 | 1400 | 0.4129 | 0.4449 |
72
- | 0.4603 | 6.6964 | 1500 | 0.4046 | 0.4255 |
73
- | 0.4229 | 7.1429 | 1600 | 0.3939 | 0.4150 |
74
- | 0.455 | 7.5893 | 1700 | 0.4133 | 0.4155 |
75
- | 0.4501 | 8.0357 | 1800 | 0.3978 | 0.4065 |
76
- | 0.45 | 8.4821 | 1900 | 0.3925 | 0.4231 |
77
- | 0.4226 | 8.9286 | 2000 | 0.3901 | 0.4098 |
78
- | 0.3973 | 9.375 | 2100 | 0.3810 | 0.4056 |
79
- | 0.4038 | 9.8214 | 2200 | 0.4178 | 0.4117 |
80
- | 0.4559 | 10.2679 | 2300 | 0.3875 | 0.4075 |
81
- | 0.4399 | 10.7143 | 2400 | 0.3742 | 0.3990 |
82
- | 0.3545 | 11.1607 | 2500 | 0.3818 | 0.4013 |
83
- | 0.4452 | 11.6071 | 2600 | 0.3906 | 0.3980 |
84
- | 0.4014 | 12.0536 | 2700 | 0.3752 | 0.3999 |
85
 
86
 
87
  ### Framework versions
 
3
  license: cc-by-nc-4.0
4
  base_model: facebook/mms-1b-all
5
  tags:
 
 
 
6
  - generated_from_trainer
7
  metrics:
8
  - wer
 
16
 
17
  # mms-1b-toigen-balanced-model
18
 
19
+ This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.3314
22
+ - Wer: 0.3856
23
 
24
  ## Model description
25
 
 
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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
  - lr_scheduler_warmup_steps: 100
50
+ - num_epochs: 2500.0
51
  - mixed_precision_training: Native AMP
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Wer |
56
  |:-------------:|:-------:|:----:|:---------------:|:------:|
57
+ | 14.2297 | 0.8850 | 100 | 3.4836 | 1.0056 |
58
+ | 4.1389 | 1.7699 | 200 | 0.5562 | 0.5694 |
59
+ | 1.3643 | 2.6549 | 300 | 0.4360 | 0.4958 |
60
+ | 1.1715 | 3.5398 | 400 | 0.3980 | 0.4824 |
61
+ | 1.1309 | 4.4248 | 500 | 0.3785 | 0.4583 |
62
+ | 1.0283 | 5.3097 | 600 | 0.3741 | 0.4477 |
63
+ | 1.0148 | 6.1947 | 700 | 0.3669 | 0.4403 |
64
+ | 0.9961 | 7.0796 | 800 | 0.3607 | 0.4356 |
65
+ | 0.9248 | 7.9646 | 900 | 0.3581 | 0.4236 |
66
+ | 0.9482 | 8.8496 | 1000 | 0.3463 | 0.4356 |
67
+ | 0.8815 | 9.7345 | 1100 | 0.3488 | 0.4273 |
68
+ | 0.8209 | 10.6195 | 1200 | 0.3384 | 0.4 |
69
+ | 0.8754 | 11.5044 | 1300 | 0.3459 | 0.4051 |
70
+ | 0.8454 | 12.3894 | 1400 | 0.3317 | 0.3884 |
71
+ | 0.8164 | 13.2743 | 1500 | 0.3319 | 0.4032 |
72
+ | 0.7673 | 14.1593 | 1600 | 0.3311 | 0.3921 |
73
+ | 0.7953 | 15.0442 | 1700 | 0.3333 | 0.3944 |
74
+ | 0.7527 | 15.9292 | 1800 | 0.3313 | 0.3917 |
75
+ | 0.763 | 16.8142 | 1900 | 0.3278 | 0.3931 |
76
+ | 0.7319 | 17.6991 | 2000 | 0.3234 | 0.3755 |
77
+ | 0.7352 | 18.5841 | 2100 | 0.3248 | 0.3806 |
78
+ | 0.7017 | 19.4690 | 2200 | 0.3334 | 0.3852 |
79
+ | 0.6902 | 20.3540 | 2300 | 0.3304 | 0.3889 |
80
+ | 0.707 | 21.2389 | 2400 | 0.3314 | 0.3856 |
 
 
 
81
 
82
 
83
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:71077b3f507d4cd00d8228d30019f943e81d6b454a8ea34bd390276156108c3c
3
  size 3858885800
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4499c8194e3ce1b11feb69453bcb04f581a08cf8cb3d5c2202b8ec504e96ab0c
3
  size 3858885800