<!-- 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. -->
wav2vec2-5epochs-3e5
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3061
- Wer: 0.2184
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: 3e-05
- 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_steps: 500
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.0438 | 0.36 | 100 | 0.2924 | 0.2284 |
0.01 | 0.72 | 200 | 0.3055 | 0.2282 |
0.0059 | 1.08 | 300 | 0.3225 | 0.2278 |
0.0028 | 1.44 | 400 | 0.3393 | 0.2245 |
0.0039 | 1.8 | 500 | 0.3506 | 0.2247 |
0.0072 | 2.16 | 600 | 0.3605 | 0.2231 |
0.0077 | 2.52 | 700 | 0.3535 | 0.2228 |
0.0077 | 2.88 | 800 | 0.3267 | 0.2235 |
0.01 | 3.24 | 900 | 0.3117 | 0.2229 |
0.0074 | 3.6 | 1000 | 0.3061 | 0.2208 |
0.0074 | 3.96 | 1100 | 0.3086 | 0.2199 |
0.0108 | 4.32 | 1200 | 0.3074 | 0.2180 |
0.0137 | 4.68 | 1300 | 0.3061 | 0.2184 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.1
- Datasets 2.9.0
- Tokenizers 0.10.3