This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.6846 | 0.83 | 100 | 2.9086 | 1.0 |
| 2.8686 | 1.67 | 200 | 2.8922 | 1.0 |
| 2.8805 | 2.5 | 300 | 2.9326 | 1.0 |
| 2.8613 | 3.33 | 400 | 2.8698 | 1.0 |
| 2.8643 | 4.17 | 500 | 2.9027 | 1.0 |
| 2.8688 | 5.0 | 600 | 2.9544 | 1.0 |
| 2.8689 | 5.83 | 700 | 2.8914 | 1.0 |
| 2.8558 | 6.67 | 800 | 2.8762 | 1.0 |
| 2.8537 | 7.5 | 900 | 2.8982 | 1.0 |
| 2.8522 | 8.33 | 1000 | 2.8820 | 1.0 |
| 2.8468 | 9.17 | 1100 | 2.8760 | 1.0 |
| 2.8454 | 10.0 | 1200 | 2.8795 | 1.0 |
last-checkpoint/optimizer.pt 755 MBlast-checkpoint/pytorch_model.bin 378 MBpytorch_model.bin 378 MBlast-checkpoint/rng_state.pth 14.5 kBlast-checkpoint/training_args.bin 2.9 kBtraining_args.bin 2.9 kBlast-checkpoint/scheduler.pt 0.6 kB