Hugging Face's logo

makhataei
/
Whisper-Small-Common-Voice

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Persian
whisper
fa-asr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Whisper-Small-Common-Voice / README.md
makhataei's picture
makhataei
End of training
e8c5851
preview code
|
raw
history blame
No virus
7.74 kB
metadata
language:
  - fa
license: apache-2.0
base_model: makhataei/Whisper-Small-Common-Voice
tags:
  - fa-asr
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_15_0
metrics:
  - wer
model-index:
  - name: Whisper Small Persian
    results: []

Whisper Small Persian

This model is a fine-tuned version of makhataei/Whisper-Small-Common-Voice on the Common Voice 15.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9637
  • Wer: 53.0122

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-06
  • train_batch_size: 10
  • eval_batch_size: 10
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 40
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 10000

Training results

Training Loss Epoch Step Validation Loss Wer
0.0003 0.14 100 0.8127 50.1960
0.0003 0.28 200 0.8106 50.8591
0.0003 0.42 300 0.8138 50.2935
0.0005 0.56 400 0.8216 51.2345
0.0003 0.7 500 0.8295 50.0918
0.0003 0.83 600 0.8331 53.4124
0.0003 0.97 700 0.8269 54.9288
0.0003 1.11 800 0.8295 51.0178
0.0005 1.25 900 0.8341 50.1149
0.0007 1.39 1000 0.8423 51.9819
0.0005 1.53 1100 0.8324 52.3706
0.0003 1.67 1200 0.8411 51.8662
0.0002 1.81 1300 0.8545 52.8402
0.0004 1.95 1400 0.8619 54.0242
0.0002 2.09 1500 0.8556 54.8296
0.0004 2.23 1600 0.8291 53.9581
0.0003 2.36 1700 0.8633 51.2047
0.0003 2.5 1800 0.8557 53.7249
0.0005 2.64 1900 0.8551 51.7190
0.0003 2.78 2000 0.8418 52.9030
0.0002 2.92 2100 0.8522 50.9467
0.0002 3.06 2200 0.8798 51.2047
0.0003 3.2 2300 0.8545 51.4395
0.0002 3.34 2400 0.8633 51.0212
0.0007 3.48 2500 0.8644 53.8440
0.0002 3.62 2600 0.8598 52.5029
0.0002 3.76 2700 0.8578 52.0679
0.0002 3.89 2800 0.8672 52.1027
0.0001 4.03 2900 0.8655 52.3706
0.0001 4.17 3000 0.8741 52.2350
0.0001 4.31 3100 0.8716 53.0056
0.0001 4.45 3200 0.8758 51.0327
0.0005 4.59 3300 0.8636 51.8662
0.0001 4.73 3400 0.8725 51.0807
0.0001 4.87 3500 0.8781 51.1700
0.0001 5.01 3600 0.8806 50.7450
0.0001 5.15 3700 0.8835 50.6210
0.0001 5.29 3800 0.8852 51.1121
0.0001 5.42 3900 0.8874 51.1700
0.0001 5.56 4000 0.8894 51.3998
0.0002 5.7 4100 0.8899 51.4246
0.0001 5.84 4200 0.8927 51.6992
0.0001 5.98 4300 0.8933 51.8993
0.0001 6.12 4400 0.8966 51.7835
0.0001 6.26 4500 0.8980 51.8381
0.0001 6.4 4600 0.8973 51.7107
0.0001 6.54 4700 0.9008 51.5553
0.0001 6.68 4800 0.9029 51.1220
0.0001 6.82 4900 0.9030 51.3221
0.0001 6.95 5000 0.9039 52.1605
0.0001 7.09 5100 0.9084 52.1440
0.0001 7.23 5200 0.9106 51.9505
0.0001 7.37 5300 0.9117 52.6219
0.0001 7.51 5400 0.9133 52.4830
0.0002 7.65 5500 0.9187 51.3320
0.0001 7.79 5600 0.9184 52.3954
0.0001 7.93 5700 0.9185 52.5392
0.0001 8.07 5800 0.9209 53.1263
0.0001 8.21 5900 0.9232 53.0965
0.0001 8.34 6000 0.9242 53.6737
0.0001 8.48 6100 0.9220 52.6996
0.0001 8.62 6200 0.9228 52.6500
0.0001 8.76 6300 0.9255 52.3838
0.0001 8.9 6400 0.9269 53.0138
0.0001 9.04 6500 0.9298 52.9345
0.0001 9.18 6600 0.9317 53.2222
0.0001 9.32 6700 0.9337 53.1974
0.0001 9.46 6800 0.9354 52.9130
0.0001 9.6 6900 0.9379 52.8865
0.0001 9.74 7000 0.9407 52.9560
0.0001 9.87 7100 0.9399 52.5045
0.0001 10.01 7200 0.9394 52.9113
0.0001 10.15 7300 0.9423 52.9064
0.0001 10.29 7400 0.9422 52.9477
0.0001 10.43 7500 0.9445 53.2305
0.0001 10.57 7600 0.9452 53.1842
0.0001 10.71 7700 0.9478 53.3562
0.0001 10.85 7800 0.9451 52.9113
0.0001 10.99 7900 0.9476 52.6616
0.0 11.13 8000 0.9502 52.3606
0.0 11.27 8100 0.9518 52.7294
0.0 11.4 8200 0.9523 52.8799
0.0 11.54 8300 0.9540 52.8419
0.0001 11.68 8400 0.9542 53.0486
0.0 11.82 8500 0.9569 53.0453
0.0 11.96 8600 0.9576 52.9576
0.0 12.1 8700 0.9589 53.2371
0.0 12.24 8800 0.9599 53.2057
0.0 12.38 8900 0.9605 53.3165
0.0 12.52 9000 0.9603 52.9576
0.0 12.66 9100 0.9608 52.5789
0.0 12.8 9200 0.9609 53.2288
0.0 12.93 9300 0.9611 53.1759
0.0 13.07 9400 0.9618 53.1296
0.0001 13.21 9500 0.9632 53.0618
0.0 13.35 9600 0.9632 52.9593
0.0 13.49 9700 0.9633 52.9923
0.0 13.63 9800 0.9635 53.1379
0.0 13.77 9900 0.9637 53.0122
0.0 13.91 10000 0.9637 53.0122

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.15.0
  • Tokenizers 0.15.0