Hugging Face's logo

DewiBrynJones
/
wav2vec2-xlsr-53-ft-btb-ccv-cy

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
DewiBrynJones commited on
Commit
273a85a
•
1 Parent(s): db2805e

Training in progress, step 1000

Browse files
Files changed (6) hide show
  1. README.md +34 -37
  2. config.json +1 -0
  3. model.safetensors +1 -1
  4. runs/Aug29_06-17-59_6d77da852b30/events.out.tfevents.1724909595.6d77da852b30.617.0 +2 -2
  5. runs/Aug29_13-43-48_6d77da852b30/events.out.tfevents.1724936574.6d77da852b30.784.0 +3 -0
  6. training_args.bin +1 -1
README.md CHANGED
@@ -1,9 +1,6 @@
1
  ---
2
- license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
5
- - automatic-speech-recognition
6
- - DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
7
  - generated_from_trainer
8
  metrics:
9
  - wer
@@ -17,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
19
 
20
- This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 1.8413
23
- - Wer: 0.9956
24
 
25
  ## Model description
26
 
@@ -41,7 +38,7 @@ More information needed
41
  The following hyperparameters were used during training:
42
  - learning_rate: 0.0003
43
  - train_batch_size: 16
44
- - eval_batch_size: 64
45
  - seed: 42
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
@@ -53,36 +50,36 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:------:|:----:|:---------------:|:------:|
56
- | No log | 0.0672 | 200 | 3.0805 | 1.0 |
57
- | No log | 0.1344 | 400 | 2.9226 | 1.0 |
58
- | 4.4646 | 0.2016 | 600 | 1.3392 | 0.8646 |
59
- | 4.4646 | 0.2688 | 800 | 1.0115 | 0.7168 |
60
- | 1.1042 | 0.3360 | 1000 | 0.9009 | 0.6767 |
61
- | 1.1042 | 0.4032 | 1200 | 0.8226 | 0.5948 |
62
- | 1.1042 | 0.4704 | 1400 | 0.8068 | 0.5977 |
63
- | 0.8421 | 0.5376 | 1600 | 0.7186 | 0.5375 |
64
- | 0.8421 | 0.6048 | 1800 | 0.7201 | 0.5561 |
65
- | 0.7637 | 0.6720 | 2000 | 0.6747 | 0.5087 |
66
- | 0.7637 | 0.7392 | 2200 | 0.6709 | 0.5051 |
67
- | 0.7637 | 0.8065 | 2400 | 0.6484 | 0.4850 |
68
- | 0.7134 | 0.8737 | 2600 | 0.6388 | 0.4915 |
69
- | 0.7134 | 0.9409 | 2800 | 0.6279 | 0.4705 |
70
- | 0.6968 | 1.0081 | 3000 | 0.6452 | 0.4833 |
71
- | 0.6968 | 1.0753 | 3200 | 0.6439 | 0.4897 |
72
- | 0.6968 | 1.1425 | 3400 | 0.6750 | 0.5010 |
73
- | 0.667 | 1.2097 | 3600 | 0.7904 | 0.5592 |
74
- | 0.667 | 1.2769 | 3800 | 0.9480 | 0.5801 |
75
- | 0.9326 | 1.3441 | 4000 | 1.1457 | 0.6582 |
76
- | 0.9326 | 1.4113 | 4200 | 1.1183 | 0.6742 |
77
- | 0.9326 | 1.4785 | 4400 | 1.3061 | 0.7807 |
78
- | 1.2308 | 1.5457 | 4600 | 1.2299 | 0.7672 |
79
- | 1.2308 | 1.6129 | 4800 | 1.1784 | 0.7517 |
80
- | 1.2835 | 1.6801 | 5000 | 1.3398 | 0.7625 |
81
- | 1.2835 | 1.7473 | 5200 | 1.5754 | 0.8956 |
82
- | 1.2835 | 1.8145 | 5400 | 1.7280 | 0.9706 |
83
- | 1.6466 | 1.8817 | 5600 | 1.7603 | 0.9727 |
84
- | 1.6466 | 1.9489 | 5800 | 1.8200 | 0.9944 |
85
- | 1.8511 | 2.0161 | 6000 | 1.8413 | 0.9956 |
86
 
87
 
88
  ### Framework versions
 
1
  ---
 
2
  base_model: facebook/wav2vec2-large-xlsr-53
3
  tags:
 
 
4
  - generated_from_trainer
5
  metrics:
6
  - wer
 
14
 
15
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
16
 
17
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 1.6889
20
+ - Wer: 0.9480
21
 
22
  ## Model description
23
 
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 0.0003
40
  - train_batch_size: 16
41
+ - eval_batch_size: 128
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
 
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Wer |
52
  |:-------------:|:------:|:----:|:---------------:|:------:|
53
+ | No log | 0.0672 | 200 | 3.0784 | 1.0 |
54
+ | No log | 0.1344 | 400 | 2.9962 | 1.0 |
55
+ | 4.6966 | 0.2016 | 600 | 3.0380 | 1.0 |
56
+ | 4.6966 | 0.2688 | 800 | 2.9616 | 1.0 |
57
+ | 3.0067 | 0.3360 | 1000 | 2.9147 | 1.0 |
58
+ | 3.0067 | 0.4032 | 1200 | 2.9158 | 1.0 |
59
+ | 3.0067 | 0.4704 | 1400 | 2.9126 | 1.0 |
60
+ | 2.9095 | 0.5376 | 1600 | 2.8195 | 1.0 |
61
+ | 2.9095 | 0.6048 | 1800 | 2.3190 | 0.9804 |
62
+ | 2.3631 | 0.6720 | 2000 | 1.5925 | 0.9349 |
63
+ | 2.3631 | 0.7392 | 2200 | 1.3529 | 0.8784 |
64
+ | 2.3631 | 0.8065 | 2400 | 1.2185 | 0.8690 |
65
+ | 1.3164 | 0.8737 | 2600 | 1.1746 | 0.8566 |
66
+ | 1.3164 | 0.9409 | 2800 | 1.2047 | 0.8444 |
67
+ | 1.2708 | 1.0081 | 3000 | 1.3975 | 0.8781 |
68
+ | 1.2708 | 1.0753 | 3200 | 1.5273 | 0.9086 |
69
+ | 1.2708 | 1.1425 | 3400 | 1.5937 | 0.9166 |
70
+ | 1.5876 | 1.2097 | 3600 | 1.4998 | 0.9331 |
71
+ | 1.5876 | 1.2769 | 3800 | 1.6366 | 0.9646 |
72
+ | 1.6623 | 1.3441 | 4000 | 1.6667 | 0.9701 |
73
+ | 1.6623 | 1.4113 | 4200 | 1.5727 | 0.9483 |
74
+ | 1.6623 | 1.4785 | 4400 | 1.6119 | 0.9611 |
75
+ | 1.6759 | 1.5457 | 4600 | 1.5941 | 0.9337 |
76
+ | 1.6759 | 1.6129 | 4800 | 1.4534 | 0.9059 |
77
+ | 1.5779 | 1.6801 | 5000 | 2.0221 | 0.9572 |
78
+ | 1.5779 | 1.7473 | 5200 | 1.7697 | 0.9399 |
79
+ | 1.5779 | 1.8145 | 5400 | 1.6657 | 0.9377 |
80
+ | 1.775 | 1.8817 | 5600 | 1.7365 | 0.9714 |
81
+ | 1.775 | 1.9489 | 5800 | 1.6953 | 0.9580 |
82
+ | 1.7507 | 2.0161 | 6000 | 1.6889 | 0.9480 |
83
 
84
 
85
  ### Framework versions
config.json CHANGED
@@ -53,6 +53,7 @@
53
  "feat_proj_dropout": 0.0,
54
  "feat_quantizer_dropout": 0.0,
55
  "final_dropout": 0.0,
 
56
  "hidden_act": "gelu",
57
  "hidden_dropout": 0.0,
58
  "hidden_size": 1024,
 
53
  "feat_proj_dropout": 0.0,
54
  "feat_quantizer_dropout": 0.0,
55
  "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
  "hidden_act": "gelu",
58
  "hidden_dropout": 0.0,
59
  "hidden_size": 1024,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bd3bfb7e107ce621e59d3521c2d26d15139650539b0f8edceb60a95f09ee13bb
3
  size 1262004280
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8beb8cdf9fb4fd709542efb4492a39893107886ff9277a83766dbf50b8284f6e
3
  size 1262004280
runs/Aug29_06-17-59_6d77da852b30/events.out.tfevents.1724909595.6d77da852b30.617.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8cd295f4f22531f7ed09d69b808d4e78edaaf4282b396e45d00251380e4f418
3
- size 16728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:533fecc8009382bdb7dbdec83ef6aefe92a9db279abd4d36bed40c47f28742cf
3
+ size 19094
runs/Aug29_13-43-48_6d77da852b30/events.out.tfevents.1724936574.6d77da852b30.784.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f687967cfd66d3541a07a2128891452005077772116b5d50861138fe8ebcda23
3
+ size 8717
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b28cc66d4b54fcb16e48762e4cc512b3ba89ed501f58a67004c3299ff7bf0e66
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8063efb054ec3e93bcdcba3cf32e816722078067e7479101e2c87deec9ee0843
3
  size 5240