---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-xls-r-300m-Arabic
  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-large-xls-r-300m-Arabic

This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0136
- Wer: 0.0322
- Cer: 0.0110

## 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: 0.0005
- train_batch_size: 16
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- num_epochs: 30.0
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
| 11.9559       | 1.0   | 51   | 4.0498          | 1.0    | 1.0    |
| 3.4291        | 2.0   | 102  | 3.1451          | 1.0    | 1.0    |
| 3.1695        | 3.0   | 153  | 3.1403          | 1.0    | 1.0    |
| 3.1385        | 4.0   | 204  | 3.0922          | 1.0    | 1.0    |
| 3.0936        | 5.0   | 255  | 3.0580          | 1.0    | 1.0    |
| 3.0079        | 6.0   | 306  | 2.8385          | 1.0    | 1.0    |
| 2.6483        | 7.0   | 357  | 2.1765          | 1.0    | 0.9806 |
| 1.6589        | 8.0   | 408  | 0.7955          | 0.8652 | 0.4629 |
| 0.5616        | 9.0   | 459  | 0.1553          | 0.2185 | 0.0561 |
| 0.2056        | 10.0  | 510  | 0.0578          | 0.0867 | 0.0225 |
| 0.1104        | 11.0  | 561  | 0.0485          | 0.0881 | 0.0277 |
| 0.0903        | 12.0  | 612  | 0.0345          | 0.0480 | 0.0132 |
| 0.067         | 13.0  | 663  | 0.0279          | 0.0365 | 0.0110 |
| 0.0616        | 14.0  | 714  | 0.0292          | 0.0570 | 0.0183 |
| 0.043         | 15.0  | 765  | 0.0210          | 0.0554 | 0.0199 |
| 0.0353        | 16.0  | 816  | 0.0190          | 0.0494 | 0.0173 |
| 0.0329        | 17.0  | 867  | 0.0203          | 0.0443 | 0.0147 |
| 0.0318        | 18.0  | 918  | 0.0248          | 0.0446 | 0.0160 |
| 0.0283        | 19.0  | 969  | 0.0203          | 0.0337 | 0.0100 |
| 0.0242        | 20.0  | 1020 | 0.0193          | 0.0230 | 0.0056 |
| 0.0216        | 21.0  | 1071 | 0.0185          | 0.0233 | 0.0066 |
| 0.02          | 22.0  | 1122 | 0.0159          | 0.0267 | 0.0077 |
| 0.0176        | 23.0  | 1173 | 0.0165          | 0.0244 | 0.0078 |
| 0.0184        | 24.0  | 1224 | 0.0171          | 0.0298 | 0.0090 |
| 0.0146        | 25.0  | 1275 | 0.0154          | 0.0344 | 0.0129 |
| 0.0126        | 26.0  | 1326 | 0.0156          | 0.0337 | 0.0119 |
| 0.0135        | 27.0  | 1377 | 0.0143          | 0.0354 | 0.0121 |
| 0.0148        | 28.0  | 1428 | 0.0136          | 0.0313 | 0.0110 |
| 0.011         | 29.0  | 1479 | 0.0140          | 0.0324 | 0.0107 |
| 0.0107        | 30.0  | 1530 | 0.0136          | 0.0322 | 0.0110 |


### Framework versions

- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2