wav2vec2-base-960h-speech-commands-v1
This model is a fine-tuned version of facebook/wav2vec2-base-960h on the speech_commands dataset. It achieves the following results on the evaluation set:
- Loss: 3.5839
- Accuracy: 0.1727
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: 48
- eval_batch_size: 48
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 192
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.0 | 1.0 | 206 | 3.5839 | 0.1727 |
| 0.0 | 2.0 | 412 | 3.5839 | 0.1727 |
| 0.0 | 3.0 | 618 | 3.5839 | 0.1727 |
| 0.0 | 4.0 | 824 | 3.5839 | 0.1727 |
| 0.0 | 5.0 | 1030 | 3.5839 | 0.1727 |
Framework versions
- Transformers 4.51.2
- Pytorch 2.6.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1