---
library_name: transformers
license: apache-2.0
base_model: alamsher/wav2vec2-large-xlsr-53-common-voice-sw
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: model
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: common_voice_17_0
      type: common_voice_17_0
      config: sw
      split: validation
      args: sw
    metrics:
    - name: Wer
      type: wer
      value: 0.18270008084074374
---

<!-- 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. -->

# model

This model is a fine-tuned version of [alamsher/wav2vec2-large-xlsr-53-common-voice-sw](https://huggingface.co/alamsher/wav2vec2-large-xlsr-53-common-voice-sw) on the common_voice_17_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2929
- Wer: 0.1827

## 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: 3e-05
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.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: 15
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step  | Validation Loss | Wer    |
|:-------------:|:-------:|:-----:|:---------------:|:------:|
| 0.2938        | 0.5028  | 500   | 0.3439          | 0.1940 |
| 0.273         | 1.0050  | 1000  | 0.3088          | 0.1923 |
| 0.2327        | 1.5078  | 1500  | 0.3031          | 0.1907 |
| 0.2351        | 2.0101  | 2000  | 0.2877          | 0.1897 |
| 0.2213        | 2.5128  | 2500  | 0.2915          | 0.1889 |
| 0.2172        | 3.0151  | 3000  | 0.2863          | 0.1879 |
| 0.191         | 3.5178  | 3500  | 0.2881          | 0.1880 |
| 0.2048        | 4.0201  | 4000  | 0.2832          | 0.1875 |
| 0.1928        | 4.5229  | 4500  | 0.2825          | 0.1863 |
| 0.1871        | 5.0251  | 5000  | 0.2861          | 0.1863 |
| 0.1856        | 5.5279  | 5500  | 0.2856          | 0.1856 |
| 0.1763        | 6.0302  | 6000  | 0.2854          | 0.1856 |
| 0.1707        | 6.5329  | 6500  | 0.2883          | 0.1853 |
| 0.1714        | 7.0352  | 7000  | 0.2849          | 0.1850 |
| 0.1577        | 7.5380  | 7500  | 0.2875          | 0.1851 |
| 0.162         | 8.0402  | 8000  | 0.2852          | 0.1850 |
| 0.1489        | 8.5430  | 8500  | 0.2911          | 0.1833 |
| 0.1674        | 9.0452  | 9000  | 0.2887          | 0.1826 |
| 0.1818        | 9.5480  | 9500  | 0.2894          | 0.1828 |
| 0.165         | 10.0503 | 10000 | 0.2915          | 0.1830 |
| 0.1715        | 10.5530 | 10500 | 0.2929          | 0.1827 |


### Framework versions

- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2