Hugging Face's logo

EzraWilliam
/
wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod18

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
EzraWilliam commited on
Commit
c9ffc61
•
1 Parent(s): 7837b0c

End of training

Browse files
Files changed (3) hide show
  1. README.md +15 -21
  2. model.safetensors +1 -1
  3. runs/Apr27_08-35-58_b4b3371fdd0f/events.out.tfevents.1714207086.b4b3371fdd0f.171.0 +2 -2
README.md CHANGED
@@ -1,8 +1,8 @@
1
  ---
2
  license: apache-2.0
 
3
  tags:
4
  - generated_from_trainer
5
- base_model: facebook/wav2vec2-large-xlsr-53
6
  datasets:
7
  - common_voice_13_0
8
  metrics:
@@ -11,8 +11,8 @@ model-index:
11
  - name: wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod18
12
  results:
13
  - task:
14
- type: automatic-speech-recognition
15
  name: Automatic Speech Recognition
 
16
  dataset:
17
  name: common_voice_13_0
18
  type: common_voice_13_0
@@ -20,9 +20,9 @@ model-index:
20
  split: test
21
  args: id
22
  metrics:
23
- - type: wer
24
- value: 0.4212758112094395
25
- name: Wer
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice_13_0 dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.4870
36
- - Wer: 0.4213
37
 
38
  ## Model description
39
 
@@ -53,30 +53,24 @@ More information needed
53
 
54
  The following hyperparameters were used during training:
55
  - learning_rate: 0.0001
56
- - train_batch_size: 16
57
  - eval_batch_size: 8
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
- - num_epochs: 12
62
  - mixed_precision_training: Native AMP
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Wer |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|
68
- | 2.8991 | 1.0 | 278 | 2.8268 | 1.0 |
69
- | 0.7566 | 2.0 | 556 | 0.6085 | 0.6369 |
70
- | 0.4602 | 3.0 | 834 | 0.4967 | 0.5786 |
71
- | 0.3297 | 4.0 | 1112 | 0.4746 | 0.4981 |
72
- | 0.2545 | 5.0 | 1390 | 0.4570 | 0.4814 |
73
- | 0.2098 | 6.0 | 1668 | 0.4639 | 0.4817 |
74
- | 0.1765 | 7.0 | 1946 | 0.4796 | 0.4663 |
75
- | 0.1568 | 8.0 | 2224 | 0.4876 | 0.4485 |
76
- | 0.1514 | 9.0 | 2502 | 0.4651 | 0.4214 |
77
- | 0.131 | 10.0 | 2780 | 0.4682 | 0.4276 |
78
- | 0.1228 | 11.0 | 3058 | 0.4814 | 0.4234 |
79
- | 0.1153 | 12.0 | 3336 | 0.4870 | 0.4213 |
80
 
81
 
82
  ### Framework versions
 
1
  ---
2
  license: apache-2.0
3
+ base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
5
  - generated_from_trainer
 
6
  datasets:
7
  - common_voice_13_0
8
  metrics:
 
11
  - name: wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod18
12
  results:
13
  - task:
 
14
  name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
  dataset:
17
  name: common_voice_13_0
18
  type: common_voice_13_0
 
20
  split: test
21
  args: id
22
  metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 0.592551622418879
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice_13_0 dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.5657
36
+ - Wer: 0.5926
37
 
38
  ## Model description
39
 
 
53
 
54
  The following hyperparameters were used during training:
55
  - learning_rate: 0.0001
56
+ - train_batch_size: 32
57
  - eval_batch_size: 8
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
+ - num_epochs: 6
62
  - mixed_precision_training: Native AMP
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Wer |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|
68
+ | 4.9488 | 1.0 | 139 | 2.9365 | 1.0 |
69
+ | 2.9074 | 2.0 | 278 | 2.8074 | 1.0 |
70
+ | 2.5174 | 3.0 | 417 | 1.2338 | 0.9646 |
71
+ | 1.0341 | 4.0 | 556 | 0.6692 | 0.6851 |
72
+ | 0.6877 | 5.0 | 695 | 0.5818 | 0.6041 |
73
+ | 0.492 | 6.0 | 834 | 0.5657 | 0.5926 |
 
 
 
 
 
 
74
 
75
 
76
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9c852fc20d7c27a20269d42df45d509ee35e92a84d46a475505462b0043d6f5
3
  size 1261991980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8da567faa0d6b991c5d5bb664d5cc4601d99673af06548000489c207d8173955
3
  size 1261991980
runs/Apr27_08-35-58_b4b3371fdd0f/events.out.tfevents.1714207086.b4b3371fdd0f.171.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac30d0be7ef1eae1908e21c389fb8fb0a0c7e645323a78babc31765fcaab2cd8
3
- size 8468
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b57a644d0acc013972ec6e15b3865c745e4611bb4e365166b4c43b76a548fd6
3
+ size 10409