wav2vec2-large-xlsr-53-english-finetuned-ravdess-v6
This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-english on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0166
- Accuracy: 0.625
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.08 | 0.15 | 25 | 2.0652 | 0.125 |
| 2.057 | 0.31 | 50 | 2.0288 | 0.2083 |
| 2.0031 | 0.46 | 75 | 2.0556 | 0.1667 |
| 1.958 | 0.62 | 100 | 1.9128 | 0.2222 |
| 1.726 | 0.77 | 125 | 1.7048 | 0.3681 |
| 1.6563 | 0.93 | 150 | 1.6522 | 0.3542 |
| 1.7092 | 1.08 | 175 | 1.7439 | 0.2986 |
| 1.5645 | 1.23 | 200 | 1.5394 | 0.4236 |
| 1.4945 | 1.39 | 225 | 1.3462 | 0.5069 |
| 1.4193 | 1.54 | 250 | 1.3745 | 0.4514 |
| 1.3488 | 1.7 | 275 | 1.2707 | 0.5208 |
| 1.3205 | 1.85 | 300 | 1.3819 | 0.5278 |
| 1.2814 | 2.01 | 325 | 1.2694 | 0.5556 |
| 1.118 | 2.16 | 350 | 1.1216 | 0.5625 |
| 1.0507 | 2.31 | 375 | 1.0795 | 0.6042 |
| 0.9967 | 2.47 | 400 | 1.1243 | 0.5764 |
| 0.9471 | 2.62 | 425 | 1.0740 | 0.6181 |
| 0.8582 | 2.78 | 450 | 1.0422 | 0.5903 |
| 0.9264 | 2.93 | 475 | 1.0178 | 0.625 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3