---
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
- automatic-speech-recognition
- DewiBrynJones/banc-trawsgrifiadau-bangor-normalized
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-xlsr-53-ft-btb-cy
  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-xlsr-53-ft-btb-cy

This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-NORMALIZED - DEFAULT dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4086
- Wer: 0.3122

## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 2600
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| No log        | 0.1414 | 100  | 3.6264          | 1.0    |
| No log        | 0.2829 | 200  | 3.1080          | 1.0    |
| No log        | 0.4243 | 300  | 2.5440          | 0.9863 |
| No log        | 0.5658 | 400  | 1.2381          | 0.8110 |
| 3.4586        | 0.7072 | 500  | 1.0036          | 0.7184 |
| 3.4586        | 0.8487 | 600  | 0.8524          | 0.6379 |
| 3.4586        | 0.9901 | 700  | 0.7411          | 0.5673 |
| 3.4586        | 1.1315 | 800  | 0.6203          | 0.4838 |
| 3.4586        | 1.2730 | 900  | 0.6017          | 0.4820 |
| 0.8146        | 1.4144 | 1000 | 0.5637          | 0.4462 |
| 0.8146        | 1.5559 | 1100 | 0.5381          | 0.4208 |
| 0.8146        | 1.6973 | 1200 | 0.5155          | 0.4043 |
| 0.8146        | 1.8388 | 1300 | 0.4858          | 0.3903 |
| 0.8146        | 1.9802 | 1400 | 0.4758          | 0.3823 |
| 0.6294        | 2.1216 | 1500 | 0.4604          | 0.3642 |
| 0.6294        | 2.2631 | 1600 | 0.4531          | 0.3536 |
| 0.6294        | 2.4045 | 1700 | 0.4435          | 0.3511 |
| 0.6294        | 2.5460 | 1800 | 0.4366          | 0.3499 |
| 0.6294        | 2.6874 | 1900 | 0.4309          | 0.3451 |
| 0.4914        | 2.8289 | 2000 | 0.4252          | 0.3368 |
| 0.4914        | 2.9703 | 2100 | 0.4218          | 0.3285 |
| 0.4914        | 3.1117 | 2200 | 0.4208          | 0.3251 |
| 0.4914        | 3.2532 | 2300 | 0.4144          | 0.3236 |
| 0.4914        | 3.3946 | 2400 | 0.4140          | 0.3165 |
| 0.4011        | 3.5361 | 2500 | 0.4133          | 0.3157 |
| 0.4011        | 3.6775 | 2600 | 0.4086          | 0.3122 |


### Framework versions

- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1