<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
whisper-tiny-three-data-v1
This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2724
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: 0.0015
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- distributed_type: tpu
- num_devices: 8
- total_train_batch_size: 256
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1024
- training_steps: 52975
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.0917 | 1.0 | 2522 | 1.1759 |
0.872 | 2.0 | 5044 | 0.9783 |
0.7945 | 3.0 | 7566 | 0.9237 |
0.7433 | 4.0 | 10088 | 0.8313 |
0.7025 | 5.0 | 12610 | 0.8000 |
0.6719 | 6.0 | 15132 | 0.7773 |
0.6305 | 7.0 | 17654 | 0.7364 |
0.569 | 8.0 | 20176 | 0.6671 |
0.5253 | 9.0 | 22698 | 0.6295 |
0.4775 | 10.0 | 25220 | 0.5687 |
0.4402 | 11.0 | 27742 | 0.5376 |
0.406 | 12.0 | 30264 | 0.4973 |
0.3713 | 12.99 | 32786 | 0.4585 |
0.337 | 13.99 | 35308 | 0.4208 |
0.3053 | 14.99 | 37830 | 0.3908 |
0.2781 | 15.99 | 40352 | 0.3635 |
0.2545 | 16.99 | 42874 | 0.3384 |
0.2332 | 17.99 | 45396 | 0.3154 |
0.2121 | 18.99 | 47918 | 0.2951 |
0.1968 | 19.99 | 50440 | 0.2809 |
0.1829 | 20.99 | 52962 | 0.2725 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.1.0a0+gitcc01568
- Datasets 2.12.0
- Tokenizers 0.13.3