<!-- 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-ja-colab-new
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_10_0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.1931
- Wer: 0.2584
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
No log | 1.0 | 637 | 5.3089 | 0.9670 |
No log | 2.0 | 1274 | 3.2716 | 0.6123 |
No log | 3.0 | 1911 | 2.1797 | 0.4708 |
No log | 4.0 | 2548 | 1.8331 | 0.4113 |
6.3938 | 5.0 | 3185 | 1.5111 | 0.3460 |
6.3938 | 6.0 | 3822 | 1.3575 | 0.3132 |
6.3938 | 7.0 | 4459 | 1.2946 | 0.2957 |
6.3938 | 8.0 | 5096 | 1.2346 | 0.2762 |
1.023 | 9.0 | 5733 | 1.2053 | 0.2653 |
1.023 | 10.0 | 6370 | 1.1931 | 0.2584 |
Framework versions
- Transformers 4.21.2
- Pytorch 1.10.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1