<!-- 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_fleurs_ch_v2 - tinglin
This model is a fine-tuned version of openai/whisper-tiny on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.5884
- Cer: 33.5338
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: 3.75e-05
- train_batch_size: 32
- eval_batch_size: 8
- 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: 4000
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.1248 | 4.9 | 500 | 0.4761 | 113.1722 |
| 0.0107 | 9.8 | 1000 | 0.5241 | 66.8766 |
| 0.0034 | 14.71 | 1500 | 0.5499 | 34.3759 |
| 0.0013 | 19.61 | 2000 | 0.5678 | 41.2857 |
| 0.0006 | 24.51 | 2500 | 0.5734 | 34.4396 |
| 0.0005 | 29.41 | 3000 | 0.5817 | 33.1977 |
| 0.0004 | 34.31 | 3500 | 0.5864 | 33.6010 |
| 0.0004 | 39.22 | 4000 | 0.5884 | 33.5338 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.0
