Hugging Face's logo

gweltou
/
wav2vec2-xls-r-300m-br

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Breton
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
gweltou commited on
Commit
61f2f82
•
1 Parent(s): 7e4bcd9

Training in progress, step 1000

Browse files
Files changed (5) hide show
  1. README.md +69 -41
  2. model.safetensors +1 -1
  3. runs/Jun10_10-20-46_gweltaz-NUC10i7FNK/events.out.tfevents.1718007828.gweltaz-NUC10i7FNK.2838.0 +2 -2
  4. runs/Jun15_17-30-45_gweltaz-NUC10i7FNK/events.out.tfevents.1718465580.gweltaz-NUC10i7FNK.2297.0 +3 -0
  5. training_args.bin +1 -1
README.md CHANGED
@@ -1,29 +1,40 @@
1
  ---
2
- language:
3
- - br
4
  license: apache-2.0
 
5
  tags:
6
  - generated_from_trainer
7
- base_model: facebook/wav2vec2-xls-r-300m
8
  datasets:
9
- - mozilla-foundation/common_voice_15_0
10
  metrics:
11
  - wer
12
  model-index:
13
- - name: wav2vec2-xls-r-300m-breton
14
- results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  ---
16
 
17
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
  should probably proofread and complete it, then remove this comment. -->
19
 
20
- # wav2vec2-xls-r-300m-breton
21
 
22
- This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the [mozilla-foundation/common_voice_15_0]() dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.9939
25
- - Wer: 0.4680
26
- - Cer: 0.1711
27
 
28
  ## Model description
29
 
@@ -42,46 +53,63 @@ More information needed
42
  ### Training hyperparameters
43
 
44
  The following hyperparameters were used during training:
45
- - learning_rate: 6e-05
46
- - train_batch_size: 16
47
  - eval_batch_size: 8
48
  - seed: 42
49
  - gradient_accumulation_steps: 2
50
- - total_train_batch_size: 32
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
- - lr_scheduler_warmup_ratio: 0.08
54
- - num_epochs: 50
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
60
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
61
- | 10.4347 | 2.56 | 250 | 3.7906 | 1.0 | 1.0 |
62
- | 3.2144 | 5.13 | 500 | 3.0622 | 1.0 | 1.0 |
63
- | 2.6193 | 7.69 | 750 | 1.6063 | 0.9122 | 0.4042 |
64
- | 1.3658 | 10.26 | 1000 | 1.0917 | 0.7061 | 0.2474 |
65
- | 0.9614 | 12.82 | 1250 | 0.9805 | 0.6384 | 0.2245 |
66
- | 0.7915 | 15.38 | 1500 | 0.9340 | 0.5989 | 0.2124 |
67
- | 0.6728 | 17.95 | 1750 | 0.9135 | 0.5735 | 0.2016 |
68
- | 0.5889 | 20.51 | 2000 | 0.9381 | 0.5443 | 0.1944 |
69
- | 0.5374 | 23.08 | 2250 | 0.9517 | 0.5372 | 0.1918 |
70
- | 0.4972 | 25.64 | 2500 | 0.9171 | 0.5134 | 0.1828 |
71
- | 0.4537 | 28.21 | 2750 | 0.9457 | 0.5043 | 0.1804 |
72
- | 0.4217 | 30.77 | 3000 | 0.9479 | 0.4957 | 0.1789 |
73
- | 0.4082 | 33.33 | 3250 | 0.9400 | 0.4906 | 0.1762 |
74
- | 0.4005 | 35.9 | 3500 | 0.9710 | 0.4881 | 0.1770 |
75
- | 0.3702 | 38.46 | 3750 | 0.9922 | 0.4792 | 0.1757 |
76
- | 0.3473 | 41.03 | 4000 | 0.9724 | 0.4760 | 0.1734 |
77
- | 0.3405 | 43.59 | 4250 | 0.9860 | 0.4726 | 0.1722 |
78
- | 0.325 | 46.15 | 4500 | 0.9888 | 0.4679 | 0.1705 |
79
- | 0.3193 | 48.72 | 4750 | 0.9939 | 0.4680 | 0.1711 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
83
 
84
- - Transformers 4.35.2
85
- - Pytorch 2.1.0+cu121
86
- - Datasets 2.16.0
87
- - Tokenizers 0.15.0
 
1
  ---
 
 
2
  license: apache-2.0
3
+ base_model: facebook/wav2vec2-xls-r-300m
4
  tags:
5
  - generated_from_trainer
 
6
  datasets:
7
+ - common_voice_15_0
8
  metrics:
9
  - wer
10
  model-index:
11
+ - name: wav2vec2-xls-r-300m-br
12
+ results:
13
+ - task:
14
+ name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
+ dataset:
17
+ name: common_voice_15_0
18
+ type: common_voice_15_0
19
+ config: br
20
+ split: None
21
+ args: br
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 50.08524001794527
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
+ # wav2vec2-xls-r-300m-br
32
 
33
+ 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_15_0 dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.8404
36
+ - Wer: 50.0852
37
+ - Cer: 17.4519
38
 
39
  ## Model description
40
 
 
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
+ - learning_rate: 4e-05
57
+ - train_batch_size: 8
58
  - eval_batch_size: 8
59
  - seed: 42
60
  - gradient_accumulation_steps: 2
61
+ - total_train_batch_size: 16
62
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
  - lr_scheduler_type: linear
64
+ - lr_scheduler_warmup_steps: 300
65
+ - num_epochs: 40
66
  - mixed_precision_training: Native AMP
67
 
68
  ### Training results
69
 
70
+ | Training Loss | Epoch | Step | Cer | Validation Loss | Wer |
71
+ |:-------------:|:-----:|:-----:|:-------:|:---------------:|:-------:|
72
+ | 6.6871 | 1.09 | 500 | 100.0 | 3.2774 | 100.0 |
73
+ | 3.0612 | 2.18 | 1000 | 99.9339 | 2.7879 | 99.9910 |
74
+ | 1.7934 | 3.27 | 1500 | 29.4362 | 1.1762 | 80.5922 |
75
+ | 1.0914 | 4.36 | 2000 | 25.0591 | 0.9210 | 70.7941 |
76
+ | 0.8895 | 5.45 | 2500 | 23.6321 | 0.8364 | 67.1243 |
77
+ | 0.7831 | 6.54 | 3000 | 22.4169 | 0.7813 | 63.9480 |
78
+ | 0.697 | 7.63 | 3500 | 21.4625 | 0.7820 | 61.8214 |
79
+ | 0.6474 | 8.71 | 4000 | 20.7367 | 0.7471 | 59.4437 |
80
+ | 0.5969 | 9.8 | 4500 | 20.0072 | 0.7255 | 57.8914 |
81
+ | 0.5677 | 10.89 | 5000 | 20.0563 | 0.7440 | 57.5774 |
82
+ | 0.5286 | 11.98 | 5500 | 19.7483 | 0.7622 | 56.2494 |
83
+ | 0.5054 | 13.07 | 6000 | 19.1510 | 0.7318 | 55.1548 |
84
+ | 0.4831 | 14.16 | 6500 | 19.2096 | 0.7731 | 54.6882 |
85
+ | 0.4606 | 15.25 | 7000 | 19.0282 | 0.7457 | 54.4459 |
86
+ | 0.4432 | 16.34 | 7500 | 18.9923 | 0.7638 | 54.1319 |
87
+ | 0.4116 | 17.43 | 8000 | 18.6880 | 0.7576 | 53.3692 |
88
+ | 0.4099 | 18.52 | 8500 | 18.6653 | 0.7944 | 53.1359 |
89
+ | 0.3991 | 19.61 | 9000 | 18.7258 | 0.8229 | 52.9296 |
90
+ | 0.3796 | 20.7 | 9500 | 18.4555 | 0.8106 | 52.3194 |
91
+ | 0.3715 | 21.79 | 10000 | 18.1078 | 0.7611 | 51.8798 |
92
+ | 0.359 | 22.88 | 10500 | 18.4139 | 0.7921 | 52.2207 |
93
+ | 0.3384 | 23.97 | 11000 | 18.0624 | 0.8022 | 51.4850 |
94
+ | 0.3367 | 25.05 | 11500 | 0.7921 | 51.5209 | 18.0322 |
95
+ | 0.3295 | 26.14 | 12000 | 0.8354 | 51.4491 | 17.9811 |
96
+ | 0.3183 | 27.23 | 12500 | 0.8171 | 51.0991 | 17.8488 |
97
+ | 0.3135 | 28.32 | 13000 | 0.8094 | 50.9915 | 17.7354 |
98
+ | 0.309 | 29.41 | 13500 | 0.8632 | 50.8659 | 17.7978 |
99
+ | 0.2922 | 30.5 | 14000 | 0.8268 | 50.7672 | 17.6636 |
100
+ | 0.2987 | 31.59 | 14500 | 0.8108 | 50.2557 | 17.5918 |
101
+ | 0.2914 | 32.68 | 15000 | 0.8237 | 50.0224 | 17.4708 |
102
+ | 0.2893 | 33.77 | 15500 | 0.8450 | 50.1211 | 17.3877 |
103
+ | 0.2853 | 34.86 | 16000 | 0.8354 | 50.4800 | 17.5464 |
104
+ | 0.2791 | 35.95 | 16500 | 0.8424 | 50.1929 | 17.5257 |
105
+ | 0.2732 | 37.04 | 17000 | 0.8390 | 50.2826 | 17.5653 |
106
+ | 0.2691 | 38.13 | 17500 | 0.8420 | 50.1122 | 17.4671 |
107
+ | 0.2702 | 39.22 | 18000 | 0.8404 | 50.0852 | 17.4519 |
108
 
109
 
110
  ### Framework versions
111
 
112
+ - Transformers 4.39.1
113
+ - Pytorch 2.0.1+cu117
114
+ - Datasets 2.18.0
115
+ - Tokenizers 0.15.2
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a9536b8ded7490a374aebb1d52f04738d69c3329a56fb08db84af9deb9c1bbb2
3
  size 1262012432
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ecb966775959b98c23b98f542135a948768df64b7cf5e35e78de8831509b6c3
3
  size 1262012432
runs/Jun10_10-20-46_gweltaz-NUC10i7FNK/events.out.tfevents.1718007828.gweltaz-NUC10i7FNK.2838.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3bc1c7854c8a57802b47c05b606a5be53ef5d5aa1c59d5c57c9f0278a6211e1
3
- size 8405
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9a3f7785147324422d9cb506680dd7bff5edcbf39986c96947e40d5004ee57e
3
+ size 14569
runs/Jun15_17-30-45_gweltaz-NUC10i7FNK/events.out.tfevents.1718465580.gweltaz-NUC10i7FNK.2297.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33a09b14419b458c7f00aa72d12121186b6406b28a94cf9879990494fe33ee23
3
+ size 6677
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7d4b37fd138aa406f2bd61fababdd575f44b1c0d7018201c4298c8863c1015a6
3
  size 4539
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a317982ad2d3adbd83b66fec5d1d60e8ab09afa8e61c4f5627913f402e9085b
3
  size 4539