<!-- 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. -->
xlsr-53-1030
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.9547
- Cer: 0.3248
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1500
- num_epochs: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 6.1214 | 24.39 | 500 | 5.4835 | 0.9790 |
| 3.8554 | 48.78 | 1000 | 3.9310 | 0.9602 |
| 0.9406 | 73.17 | 1500 | 1.0191 | 0.3613 |
| 0.3241 | 97.56 | 2000 | 0.8803 | 0.3362 |
| 0.207 | 121.95 | 2500 | 0.9257 | 0.3287 |
| 0.1602 | 146.34 | 3000 | 0.9385 | 0.3201 |
| 0.1351 | 170.73 | 3500 | 0.9547 | 0.3248 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.2
