Hugging Face's logo

badrex
/
xlsr-128upper-sorbian

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Badr Abdullah commited on
Commit
cc75278
•
1 Parent(s): 13591b8

Model save

Browse files
Files changed (1) hide show
  1. README.md +31 -18
README.md CHANGED
@@ -22,20 +22,20 @@ model-index:
22
  metrics:
23
  - name: Wer
24
  type: wer
25
- value: 0.549367088607595
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
  should probably proofread and complete it, then remove this comment. -->
30
 
31
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/badr-nlp/xlsr-continual-finetuning-new/runs/xhato0mz)
32
  # xlsr-128upper-sorbian
33
 
34
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_17_0 dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.7110
37
- - Wer: 0.5494
38
- - Cer: 0.1188
39
 
40
  ## Model description
41
 
@@ -63,25 +63,38 @@ The following hyperparameters were used during training:
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
  - lr_scheduler_warmup_steps: 500
66
- - num_epochs: 50
67
  - mixed_precision_training: Native AMP
68
 
69
  ### Training results
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
72
  |:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
73
- | 3.8492 | 3.9216 | 100 | 3.9919 | 1.0 | 1.0 |
74
- | 3.1983 | 7.8431 | 200 | 3.2332 | 1.0 | 1.0 |
75
- | 2.9601 | 11.7647 | 300 | 3.0166 | 0.9873 | 0.9798 |
76
- | 0.4618 | 15.6863 | 400 | 0.7749 | 0.7557 | 0.1917 |
77
- | 0.2411 | 19.6078 | 500 | 0.7812 | 0.7013 | 0.1702 |
78
- | 0.1112 | 23.5294 | 600 | 0.7275 | 0.6405 | 0.1508 |
79
- | 0.1108 | 27.4510 | 700 | 0.7995 | 0.6247 | 0.1440 |
80
- | 0.0432 | 31.3725 | 800 | 0.7902 | 0.6139 | 0.1432 |
81
- | 0.0431 | 35.2941 | 900 | 0.7615 | 0.5797 | 0.1372 |
82
- | 0.0515 | 39.2157 | 1000 | 0.7029 | 0.5456 | 0.1234 |
83
- | 0.0241 | 43.1373 | 1100 | 0.7296 | 0.5285 | 0.1188 |
84
- | 0.0342 | 47.0588 | 1200 | 0.7110 | 0.5494 | 0.1188 |
 
 
 
 
 
 
 
 
 
 
 
 
 
85
 
86
 
87
  ### Framework versions
 
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 0.5044303797468355
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
  should probably proofread and complete it, then remove this comment. -->
30
 
31
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/badr-nlp/xlsr-continual-finetuning-new/runs/0wnfr6i1)
32
  # xlsr-128upper-sorbian
33
 
34
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_17_0 dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.7625
37
+ - Wer: 0.5044
38
+ - Cer: 0.1106
39
 
40
  ## Model description
41
 
 
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
  - lr_scheduler_warmup_steps: 500
66
+ - num_epochs: 100
67
  - mixed_precision_training: Native AMP
68
 
69
  ### Training results
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
72
  |:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
73
+ | 3.8489 | 3.9216 | 100 | 4.0479 | 1.0 | 1.0 |
74
+ | 3.1996 | 7.8431 | 200 | 3.2124 | 0.9804 | 0.9850 |
75
+ | 2.3527 | 11.7647 | 300 | 2.4026 | 1.0 | 0.6858 |
76
+ | 0.4424 | 15.6863 | 400 | 0.7917 | 0.7418 | 0.1910 |
77
+ | 0.2617 | 19.6078 | 500 | 0.7624 | 0.6804 | 0.1696 |
78
+ | 0.1421 | 23.5294 | 600 | 0.7839 | 0.6582 | 0.1579 |
79
+ | 0.097 | 27.4510 | 700 | 0.8322 | 0.6316 | 0.1495 |
80
+ | 0.0459 | 31.3725 | 800 | 0.8119 | 0.6171 | 0.1446 |
81
+ | 0.0668 | 35.2941 | 900 | 0.8534 | 0.6418 | 0.1535 |
82
+ | 0.0627 | 39.2157 | 1000 | 0.8256 | 0.6019 | 0.1397 |
83
+ | 0.0454 | 43.1373 | 1100 | 0.7747 | 0.5994 | 0.1363 |
84
+ | 0.04 | 47.0588 | 1200 | 0.8046 | 0.5810 | 0.1321 |
85
+ | 0.0563 | 50.9804 | 1300 | 0.7910 | 0.5797 | 0.1325 |
86
+ | 0.039 | 54.9020 | 1400 | 0.7370 | 0.5595 | 0.1265 |
87
+ | 0.0254 | 58.8235 | 1500 | 0.7395 | 0.5418 | 0.1188 |
88
+ | 0.0211 | 62.7451 | 1600 | 0.7582 | 0.5430 | 0.1209 |
89
+ | 0.0218 | 66.6667 | 1700 | 0.7123 | 0.5051 | 0.1121 |
90
+ | 0.0206 | 70.5882 | 1800 | 0.7912 | 0.5297 | 0.1165 |
91
+ | 0.0155 | 74.5098 | 1900 | 0.7671 | 0.5367 | 0.1183 |
92
+ | 0.0242 | 78.4314 | 2000 | 0.7926 | 0.5418 | 0.1170 |
93
+ | 0.0081 | 82.3529 | 2100 | 0.7817 | 0.5373 | 0.1221 |
94
+ | 0.0087 | 86.2745 | 2200 | 0.7989 | 0.5285 | 0.1165 |
95
+ | 0.0088 | 90.1961 | 2300 | 0.7523 | 0.5165 | 0.1141 |
96
+ | 0.0173 | 94.1176 | 2400 | 0.7646 | 0.5038 | 0.1108 |
97
+ | 0.0217 | 98.0392 | 2500 | 0.7625 | 0.5044 | 0.1106 |
98
 
99
 
100
  ### Framework versions