<!-- 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-large-xls-r-300m-italian-sofia-colab
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.7517
- Wer: 0.6195
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: 0.0003
- train_batch_size: 4
- eval_batch_size: 8
- 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: 250
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
5.4919 | 1.22 | 200 | 2.9505 | 1.0 |
2.4545 | 2.45 | 400 | 1.4653 | 0.9722 |
1.2544 | 3.67 | 600 | 0.9959 | 0.8968 |
0.8818 | 4.89 | 800 | 0.9081 | 0.7956 |
0.6625 | 6.12 | 1000 | 0.8297 | 0.7379 |
0.5166 | 7.34 | 1200 | 0.8511 | 0.7160 |
0.4189 | 8.56 | 1400 | 0.8109 | 0.7332 |
0.3641 | 9.79 | 1600 | 0.7340 | 0.6590 |
0.2836 | 11.01 | 1800 | 0.7723 | 0.6571 |
0.2296 | 12.23 | 2000 | 0.8045 | 0.6838 |
0.1962 | 13.46 | 2200 | 0.7661 | 0.6535 |
0.178 | 14.68 | 2400 | 0.7517 | 0.6195 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.10.0+cu113
- Datasets 2.10.1
- Tokenizers 0.13.2