Hugging Face's logo

sonnygeorge
/
whisper-tiny-papi

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
whisper
ASR
Papiamentu
Whisper
Speech Recognition
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
sonnygeorge commited on
Commit
e071a97
•
1 Parent(s): 5e59824

End of training

Browse files
Files changed (1) hide show
  1. README.md +27 -27
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Wer
25
  type: wer
26
- value: 85.22752415369865
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Papi ASR dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 2.0809
37
- - Wer: 85.2275
38
- - Cer: 34.1432
39
 
40
  ## Model description
41
 
@@ -60,34 +60,34 @@ The following hyperparameters were used during training:
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
- - lr_scheduler_warmup_steps: 500
64
  - training_steps: 100
65
  - mixed_precision_training: Native AMP
66
 
67
  ### Training results
68
 
69
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
70
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------:|
71
- | No log | 0.01 | 5 | 4.6530 | 101.2317 | 53.1862 |
72
- | No log | 0.01 | 10 | 4.6414 | 102.1683 | 53.8408 |
73
- | No log | 0.02 | 15 | 4.6073 | 102.0577 | 53.5251 |
74
- | No log | 0.03 | 20 | 4.5651 | 101.3275 | 52.8798 |
75
- | 4.5851 | 0.04 | 25 | 4.3546 | 100.9956 | 52.4930 |
76
- | 4.5851 | 0.04 | 30 | 4.2726 | 98.9158 | 50.6731 |
77
- | 4.5851 | 0.05 | 35 | 4.1573 | 98.3922 | 50.3358 |
78
- | 4.5851 | 0.06 | 40 | 3.9879 | 96.8360 | 48.6228 |
79
- | 4.5851 | 0.07 | 45 | 3.8552 | 96.4747 | 47.4498 |
80
- | 4.1061 | 0.07 | 50 | 3.6580 | 94.7710 | 47.1001 |
81
- | 4.1061 | 0.08 | 55 | 3.4486 | 93.1780 | 45.8250 |
82
- | 4.1061 | 0.09 | 60 | 3.2790 | 94.1441 | 46.0849 |
83
- | 4.1061 | 0.1 | 65 | 3.1437 | 93.4730 | 45.1332 |
84
- | 4.1061 | 0.1 | 70 | 3.0021 | 91.7029 | 42.6202 |
85
- | 3.2929 | 0.11 | 75 | 2.8536 | 92.1086 | 41.3095 |
86
- | 3.2929 | 0.12 | 80 | 2.7078 | 90.1468 | 38.6819 |
87
- | 3.2929 | 0.13 | 85 | 2.5575 | 88.8045 | 37.2195 |
88
- | 3.2929 | 0.13 | 90 | 2.4029 | 88.6865 | 36.6887 |
89
- | 3.2929 | 0.14 | 95 | 2.2431 | 89.0257 | 35.9986 |
90
- | 2.4377 | 0.15 | 100 | 2.0809 | 85.2275 | 34.1432 |
91
 
92
 
93
  ### Framework versions
 
23
  metrics:
24
  - name: Wer
25
  type: wer
26
+ value: 65.51368094992256
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Papi ASR dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.7807
37
+ - Wer: 65.5137
38
+ - Cer: 40.6410
39
 
40
  ## Model description
41
 
 
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
+ - lr_scheduler_warmup_steps: 10
64
  - training_steps: 100
65
  - mixed_precision_training: Native AMP
66
 
67
  ### Training results
68
 
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
70
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
71
+ | No log | 0.01 | 5 | 2.0665 | 85.5299 | 34.0333 |
72
+ | No log | 0.01 | 10 | 1.8070 | 82.8527 | 30.8192 |
73
+ | No log | 0.02 | 15 | 1.5071 | 82.2111 | 38.9217 |
74
+ | No log | 0.03 | 20 | 1.3323 | 84.4015 | 57.8317 |
75
+ | 1.6034 | 0.04 | 25 | 1.2162 | 86.8353 | 66.3752 |
76
+ | 1.6034 | 0.04 | 30 | 1.1291 | 85.2939 | 64.6158 |
77
+ | 1.6034 | 0.05 | 35 | 1.0597 | 83.7377 | 62.4571 |
78
+ | 1.6034 | 0.06 | 40 | 1.0059 | 81.0310 | 58.4058 |
79
+ | 1.6034 | 0.07 | 45 | 0.9589 | 77.0042 | 53.9104 |
80
+ | 0.9748 | 0.07 | 50 | 0.9209 | 74.7843 | 51.4639 |
81
+ | 0.9748 | 0.08 | 55 | 0.8917 | 73.7075 | 50.3048 |
82
+ | 0.9748 | 0.09 | 60 | 0.8675 | 72.9331 | 49.3238 |
83
+ | 0.9748 | 0.1 | 65 | 0.8470 | 71.5466 | 47.7252 |
84
+ | 0.9748 | 0.1 | 70 | 0.8304 | 70.7796 | 46.7612 |
85
+ | 0.8138 | 0.11 | 75 | 0.8166 | 69.7692 | 45.3762 |
86
+ | 0.8138 | 0.12 | 80 | 0.8042 | 67.9696 | 43.6446 |
87
+ | 0.8138 | 0.13 | 85 | 0.7947 | 66.9666 | 42.8322 |
88
+ | 0.8138 | 0.13 | 90 | 0.7881 | 65.6907 | 41.1903 |
89
+ | 0.8138 | 0.14 | 95 | 0.7832 | 65.5063 | 40.7261 |
90
+ | 0.7389 | 0.15 | 100 | 0.7807 | 65.5137 | 40.6410 |
91
 
92
 
93
  ### Framework versions