|
|
--- |
|
|
library_name: transformers |
|
|
license: apache-2.0 |
|
|
base_model: facebook/wav2vec2-large-xlsr-53 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- wer |
|
|
model-index: |
|
|
- name: wav2vec2-IEEEAccess-FinalRun-4Datasets |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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-IEEEAccess-FinalRun-4Datasets |
|
|
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.5364 |
|
|
- Wer: 0.2818 |
|
|
- Cer: 0.0677 |
|
|
|
|
|
## 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: 16 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 2 |
|
|
- total_train_batch_size: 32 |
|
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_ratio: 0.1 |
|
|
- num_epochs: 45 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |
|
|
|:-------------:|:-------:|:-----:|:---------------:|:------:|:------:| |
|
|
| 1.0425 | 3.2326 | 1600 | 0.4998 | 0.5122 | 0.1218 | |
|
|
| 0.7519 | 6.4651 | 3200 | 0.4336 | 0.4143 | 0.0987 | |
|
|
| 0.5768 | 9.6977 | 4800 | 0.4309 | 0.3936 | 0.0937 | |
|
|
| 0.46 | 12.9302 | 6400 | 0.3926 | 0.3480 | 0.0832 | |
|
|
| 0.382 | 16.1618 | 8000 | 0.4376 | 0.3645 | 0.0828 | |
|
|
| 0.3188 | 19.3943 | 9600 | 0.4384 | 0.3376 | 0.0798 | |
|
|
| 0.2536 | 22.6269 | 11200 | 0.4611 | 0.3110 | 0.0759 | |
|
|
| 0.2115 | 25.8595 | 12800 | 0.4612 | 0.3048 | 0.0737 | |
|
|
| 0.1882 | 29.0910 | 14400 | 0.5392 | 0.3149 | 0.0758 | |
|
|
| 0.1496 | 32.3236 | 16000 | 0.5056 | 0.2951 | 0.0728 | |
|
|
| 0.1252 | 35.5561 | 17600 | 0.5093 | 0.2924 | 0.0712 | |
|
|
| 0.1075 | 38.7887 | 19200 | 0.5203 | 0.2866 | 0.0691 | |
|
|
| 0.083 | 42.0202 | 20800 | 0.5364 | 0.2818 | 0.0677 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.52.0 |
|
|
- Pytorch 2.7.0+cu126 |
|
|
- Datasets 3.6.0 |
|
|
- Tokenizers 0.21.1 |
|
|
|
