---
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
model-index:
- name: k2e-20s_asr-scr_w2v2-base_004
  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. -->

# k2e-20s_asr-scr_w2v2-base_004

This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.6279
- Per: 0.1742
- Pcc: 0.5623
- Ctc Loss: 0.5582
- Mse Loss: 1.0308

## 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: 1e-05
- train_batch_size: 16
- eval_batch_size: 1
- seed: 1234
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2235
- training_steps: 22350
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Per    | Pcc    | Ctc Loss | Mse Loss |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:--------:|:--------:|
| 19.1543       | 3.0   | 2235  | 4.5048          | 0.9890 | 0.5904 | 3.8056   | 0.7652   |
| 4.3704        | 6.01  | 4470  | 4.4024          | 0.9890 | 0.5979 | 3.7604   | 0.8119   |
| 3.8685        | 9.01  | 6705  | 4.2418          | 0.9890 | 0.5629 | 3.5307   | 0.9394   |
| 2.968         | 12.02 | 8940  | 3.0058          | 0.5972 | 0.5624 | 1.9356   | 1.1595   |
| 1.4899        | 15.02 | 11175 | 1.8180          | 0.2433 | 0.5595 | 0.9015   | 0.9148   |
| 0.9431        | 18.02 | 13410 | 1.8391          | 0.2057 | 0.5576 | 0.6985   | 1.0942   |
| 0.7475        | 21.03 | 15645 | 1.6988          | 0.1896 | 0.5583 | 0.6179   | 1.0382   |
| 0.6357        | 24.03 | 17880 | 1.5647          | 0.1799 | 0.5494 | 0.5862   | 0.9522   |
| 0.5695        | 27.04 | 20115 | 1.6835          | 0.1758 | 0.5604 | 0.5648   | 1.0716   |
| 0.5288        | 30.04 | 22350 | 1.6279          | 0.1742 | 0.5623 | 0.5582   | 1.0308   |


### Framework versions

- Transformers 4.38.1
- Pytorch 2.0.1
- Datasets 2.18.0
- Tokenizers 0.15.2