Hugging Face's logo

cantillation
/
Teamim-small_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-08

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Hebrew
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Teamim-small_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-08 / README.md
cantillation's picture
cantillation
Model save
cc426f2 verified
preview code
|
raw
history blame
19.2 kB
metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: Teamim-small_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-08
    results: []

Teamim-small_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-08

This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1496
  • Wer: 9.8776
  • Avg Precision Exact: 0.9087
  • Avg Recall Exact: 0.9107
  • Avg F1 Exact: 0.9093
  • Avg Precision Letter Shift: 0.9225
  • Avg Recall Letter Shift: 0.9246
  • Avg F1 Letter Shift: 0.9232
  • Avg Precision Word Level: 0.9249
  • Avg Recall Word Level: 0.9267
  • Avg F1 Word Level: 0.9254
  • Avg Precision Word Shift: 0.9697
  • Avg Recall Word Shift: 0.9724
  • Avg F1 Word Shift: 0.9706
  • Precision Median Exact: 1.0
  • Recall Median Exact: 1.0
  • F1 Median Exact: 1.0
  • Precision Max Exact: 1.0
  • Recall Max Exact: 1.0
  • F1 Max Exact: 1.0
  • Precision Min Exact: 0.0
  • Recall Min Exact: 0.0
  • F1 Min Exact: 0.0
  • Precision Min Letter Shift: 0.0
  • Recall Min Letter Shift: 0.0
  • F1 Min Letter Shift: 0.0
  • Precision Min Word Level: 0.0
  • Recall Min Word Level: 0.0
  • F1 Min Word Level: 0.0
  • Precision Min Word Shift: 0.1429
  • Recall Min Word Shift: 0.1
  • F1 Min Word Shift: 0.1176

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 200000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Avg Precision Exact Avg Recall Exact Avg F1 Exact Avg Precision Letter Shift Avg Recall Letter Shift Avg F1 Letter Shift Avg Precision Word Level Avg Recall Word Level Avg F1 Word Level Avg Precision Word Shift Avg Recall Word Shift Avg F1 Word Shift Precision Median Exact Recall Median Exact F1 Median Exact Precision Max Exact Recall Max Exact F1 Max Exact Precision Min Exact Recall Min Exact F1 Min Exact Precision Min Letter Shift Recall Min Letter Shift F1 Min Letter Shift Precision Min Word Level Recall Min Word Level F1 Min Word Level Precision Min Word Shift Recall Min Word Shift F1 Min Word Shift
No log 0.0001 1 7.3549 108.0965 0.0006 0.0037 0.0009 0.0204 0.0217 0.0204 0.0078 0.0760 0.0140 0.1364 0.1552 0.1388 0.0 0.0 0.0 0.2 1.0 0.3077 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0527 0.5167 10000 0.1247 17.8011 0.8390 0.8441 0.8410 0.8596 0.8648 0.8616 0.8632 0.8682 0.8651 0.9344 0.9409 0.9369 0.9167 0.9167 0.9167 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0769 0.0667 0.0714
0.0181 1.0334 20000 0.1178 14.7550 0.8697 0.8697 0.8690 0.8884 0.8887 0.8878 0.8911 0.8918 0.8907 0.9502 0.9517 0.9501 0.9333 0.9333 0.9474 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0121 1.5501 30000 0.1202 13.1219 0.8866 0.8899 0.8878 0.9039 0.9074 0.9052 0.9067 0.9103 0.9080 0.9551 0.9595 0.9568 1.0 1.0 0.9630 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.125 0.1333
0.0041 2.0668 40000 0.1301 12.2219 0.8897 0.8925 0.8907 0.9058 0.9087 0.9068 0.9089 0.9114 0.9097 0.9579 0.9612 0.9591 1.0 1.0 0.9655 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.125 0.1 0.1176
0.0053 2.5834 50000 0.1294 12.3635 0.8863 0.8873 0.8864 0.9030 0.9042 0.9031 0.9055 0.9068 0.9057 0.9594 0.9623 0.9603 1.0 1.0 0.9630 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0909 0.1 0.0952
0.01 3.1001 60000 0.1385 12.5586 0.8824 0.8860 0.8838 0.8984 0.9023 0.8999 0.9010 0.9049 0.9025 0.9557 0.9609 0.9578 1.0 1.0 0.9630 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0037 3.6168 70000 0.1374 12.4076 0.8905 0.8936 0.8916 0.9062 0.9093 0.9073 0.9085 0.9115 0.9095 0.9561 0.9598 0.9574 1.0 1.0 0.9630 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1111 0.125
0.0031 4.1335 80000 0.1370 11.7562 0.8959 0.8971 0.8961 0.9115 0.9128 0.9117 0.9137 0.9150 0.9139 0.9610 0.9639 0.9620 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0012 4.6502 90000 0.1371 11.4573 0.8972 0.8963 0.8964 0.9126 0.9119 0.9118 0.9150 0.9146 0.9144 0.9655 0.9665 0.9655 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0021 5.1669 100000 0.1448 11.6492 0.8926 0.8952 0.8935 0.9075 0.9103 0.9085 0.9102 0.9128 0.9111 0.9601 0.9637 0.9614 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1053
0.0013 5.6836 110000 0.1440 11.0199 0.9003 0.9033 0.9014 0.9156 0.9188 0.9168 0.9179 0.9211 0.9191 0.9645 0.9685 0.9660 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0004 6.2003 120000 0.1427 10.7901 0.9033 0.9043 0.9035 0.9183 0.9194 0.9185 0.9205 0.9215 0.9206 0.9667 0.9686 0.9672 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0005 6.7170 130000 0.1434 10.7178 0.9012 0.9039 0.9021 0.9161 0.9189 0.9171 0.9185 0.9212 0.9194 0.9648 0.9684 0.9662 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1111 0.125
0.0006 7.2336 140000 0.1452 10.7776 0.9014 0.9041 0.9023 0.9162 0.9190 0.9172 0.9187 0.9211 0.9195 0.9661 0.9699 0.9676 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1111 0.125
0.0005 7.7503 150000 0.1471 10.6297 0.9037 0.9039 0.9034 0.9177 0.9180 0.9175 0.9198 0.9199 0.9195 0.9656 0.9673 0.9660 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0001 8.2670 160000 0.1484 10.3905 0.9035 0.9068 0.9048 0.9177 0.9212 0.9191 0.9200 0.9231 0.9212 0.9667 0.9703 0.9680 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0019 8.7837 170000 0.1468 10.1136 0.9078 0.9099 0.9085 0.9220 0.9242 0.9227 0.9244 0.9262 0.9249 0.9679 0.9704 0.9687 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0 9.3004 180000 0.1495 10.1167 0.9051 0.9073 0.9058 0.9194 0.9217 0.9202 0.9219 0.9241 0.9226 0.9681 0.9709 0.9691 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0 9.8171 190000 0.1497 10.0412 0.9078 0.9104 0.9087 0.9218 0.9245 0.9227 0.9243 0.9266 0.9251 0.9680 0.9713 0.9692 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176
0.0 10.3338 200000 0.1496 9.8776 0.9087 0.9107 0.9093 0.9225 0.9246 0.9232 0.9249 0.9267 0.9254 0.9697 0.9724 0.9706 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1429 0.1 0.1176

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.2.1
  • Datasets 2.20.0
  • Tokenizers 0.19.1