Hugging Face's logo

atulksingh
/
whisper-base-hi

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
English
whisper
Generated from Trainer
Eval Results
Model card Files
xet
Metrics Community
atulksingh commited on
Commit
f0a38a7
·
verified ·
1 Parent(s): 2f89b45

End of training

Browse files
Files changed (2) hide show
  1. README.md +9 -10
  2. model.safetensors +1 -1
README.md CHANGED
@@ -25,7 +25,7 @@ model-index:
25
  metrics:
26
  - name: Wer
27
  type: wer
28
- value: 2.608695652173913
29
  ---
30
 
31
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -35,8 +35,8 @@ should probably proofread and complete it, then remove this comment. -->
35
 
36
  This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Domain Based voice dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 0.0000
39
- - Wer: 2.6087
40
 
41
  ## Model description
42
 
@@ -55,26 +55,25 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 5e-06
59
  - train_batch_size: 4
60
- - eval_batch_size: 4
61
  - seed: 42
62
  - gradient_accumulation_steps: 4
63
  - total_train_batch_size: 16
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
  - lr_scheduler_warmup_steps: 100
67
- - training_steps: 2000
68
  - mixed_precision_training: Native AMP
69
 
70
  ### Training results
71
 
72
  | Training Loss | Epoch | Step | Validation Loss | Wer |
73
  |:-------------:|:-----:|:----:|:---------------:|:------:|
74
- | 0.0001 | 100.0 | 500 | 0.0001 | 1.7391 |
75
- | 0.0 | 200.0 | 1000 | 0.0001 | 1.7391 |
76
- | 0.0 | 300.0 | 1500 | 0.0000 | 2.6087 |
77
- | 0.0 | 400.0 | 2000 | 0.0000 | 2.6087 |
78
 
79
 
80
  ### Framework versions
 
25
  metrics:
26
  - name: Wer
27
  type: wer
28
+ value: 3.9215686274509802
29
  ---
30
 
31
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
35
 
36
  This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Domain Based voice dataset.
37
  It achieves the following results on the evaluation set:
38
+ - Loss: 0.0078
39
+ - Wer: 3.9216
40
 
41
  ## Model description
42
 
 
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
+ - learning_rate: 1e-06
59
  - train_batch_size: 4
60
+ - eval_batch_size: 8
61
  - seed: 42
62
  - gradient_accumulation_steps: 4
63
  - total_train_batch_size: 16
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
  - lr_scheduler_warmup_steps: 100
67
+ - training_steps: 1500
68
  - mixed_precision_training: Native AMP
69
 
70
  ### Training results
71
 
72
  | Training Loss | Epoch | Step | Validation Loss | Wer |
73
  |:-------------:|:-----:|:----:|:---------------:|:------:|
74
+ | 0.0007 | 100.0 | 500 | 0.0086 | 3.9216 |
75
+ | 0.0003 | 200.0 | 1000 | 0.0079 | 3.9216 |
76
+ | 0.0002 | 300.0 | 1500 | 0.0078 | 3.9216 |
 
77
 
78
 
79
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:df2316ca43ef16fa1fe46c8b05aae64d8422348b59b41e487636cba38ed22686
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:853ce597c3840ad5f5abfc1b7c349ef6cea20ba4729605b98f42af76bc725f7b
3
  size 290403936