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
82b06da
•
1 Parent(s): 8ac275d

Model save

Browse files
Files changed (1) hide show
  1. README.md +32 -24
README.md CHANGED
@@ -2,8 +2,6 @@
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 +15,10 @@ 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: inf
23
- - Wer: 0.3501
24
 
25
  ## Model description
26
 
@@ -46,7 +44,7 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 500
49
- - training_steps: 4000
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
@@ -54,25 +52,35 @@ The following hyperparameters were used during training:
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:------:|:----:|:---------------:|:------:|
56
  | No log | 0.0772 | 200 | inf | 1.0 |
57
- | No log | 0.1544 | 400 | inf | 0.9262 |
58
- | 3.9094 | 0.2317 | 600 | inf | 0.7767 |
59
- | 3.9094 | 0.3089 | 800 | inf | 0.7365 |
60
- | 0.9885 | 0.3861 | 1000 | inf | 0.6105 |
61
- | 0.9885 | 0.4633 | 1200 | inf | 0.5431 |
62
- | 0.9885 | 0.5405 | 1400 | inf | 0.5145 |
63
- | 0.7974 | 0.6178 | 1600 | inf | 0.5231 |
64
- | 0.7974 | 0.6950 | 1800 | inf | 0.4621 |
65
- | 0.7068 | 0.7722 | 2000 | inf | 0.4532 |
66
- | 0.7068 | 0.8494 | 2200 | inf | 0.4404 |
67
- | 0.7068 | 0.9266 | 2400 | inf | 0.4195 |
68
- | 0.6447 | 1.0039 | 2600 | inf | 0.4116 |
69
- | 0.6447 | 1.0811 | 2800 | inf | 0.3890 |
70
- | 0.5318 | 1.1583 | 3000 | inf | 0.3804 |
71
- | 0.5318 | 1.2355 | 3200 | inf | 0.3734 |
72
- | 0.5318 | 1.3127 | 3400 | inf | 0.3675 |
73
- | 0.494 | 1.3900 | 3600 | inf | 0.3590 |
74
- | 0.494 | 1.4672 | 3800 | inf | 0.3535 |
75
- | 0.4794 | 1.5444 | 4000 | inf | 0.3501 |
 
 
 
 
 
 
 
 
 
 
76
 
77
 
78
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - wer
 
15
 
16
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
17
 
18
+ 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.
19
  It achieves the following results on the evaluation set:
20
  - Loss: inf
21
+ - Wer: 0.3289
22
 
23
  ## Model description
24
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
  - lr_scheduler_warmup_steps: 500
47
+ - training_steps: 6000
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer |
53
  |:-------------:|:------:|:----:|:---------------:|:------:|
54
  | No log | 0.0772 | 200 | inf | 1.0 |
55
+ | No log | 0.1544 | 400 | inf | 0.9938 |
56
+ | 3.9317 | 0.2317 | 600 | inf | 0.7576 |
57
+ | 3.9317 | 0.3089 | 800 | inf | 0.6942 |
58
+ | 0.9699 | 0.3861 | 1000 | inf | 0.5763 |
59
+ | 0.9699 | 0.4633 | 1200 | inf | 0.5519 |
60
+ | 0.9699 | 0.5405 | 1400 | inf | 0.5174 |
61
+ | 0.8031 | 0.6178 | 1600 | inf | 0.5338 |
62
+ | 0.8031 | 0.6950 | 1800 | inf | 0.4777 |
63
+ | 0.7169 | 0.7722 | 2000 | inf | 0.4504 |
64
+ | 0.7169 | 0.8494 | 2200 | inf | 0.4500 |
65
+ | 0.7169 | 0.9266 | 2400 | inf | 0.4432 |
66
+ | 0.6687 | 1.0039 | 2600 | inf | 0.4176 |
67
+ | 0.6687 | 1.0811 | 2800 | inf | 0.4054 |
68
+ | 0.5609 | 1.1583 | 3000 | inf | 0.4009 |
69
+ | 0.5609 | 1.2355 | 3200 | inf | 0.4023 |
70
+ | 0.5609 | 1.3127 | 3400 | inf | 0.3919 |
71
+ | 0.5324 | 1.3900 | 3600 | inf | 0.3795 |
72
+ | 0.5324 | 1.4672 | 3800 | inf | 0.3752 |
73
+ | 0.5196 | 1.5444 | 4000 | inf | 0.3662 |
74
+ | 0.5196 | 1.6216 | 4200 | inf | 0.3703 |
75
+ | 0.5196 | 1.6988 | 4400 | inf | 0.3614 |
76
+ | 0.4967 | 1.7761 | 4600 | inf | 0.3530 |
77
+ | 0.4967 | 1.8533 | 4800 | inf | 0.3481 |
78
+ | 0.4735 | 1.9305 | 5000 | inf | 0.3506 |
79
+ | 0.4735 | 2.0077 | 5200 | inf | 0.3432 |
80
+ | 0.4735 | 2.0849 | 5400 | inf | 0.3369 |
81
+ | 0.4244 | 2.1622 | 5600 | inf | 0.3296 |
82
+ | 0.4244 | 2.2394 | 5800 | inf | 0.3295 |
83
+ | 0.3674 | 2.3166 | 6000 | inf | 0.3289 |
84
 
85
 
86
  ### Framework versions