---
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
- automatic-speech-recognition
- DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-xlsr-53-ft-btb-ccv-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-ccv-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-CLEAN-WITH-CCV - DEFAULT dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7821
- Wer: 0.4576

## 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: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 600
- training_steps: 6000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| No log        | 0.0631 | 200  | 3.0428          | 1.0    |
| No log        | 0.1262 | 400  | 3.1056          | 1.0    |
| 4.232         | 0.1893 | 600  | 1.2093          | 0.8286 |
| 4.232         | 0.2524 | 800  | 1.0886          | 0.7325 |
| 1.0379        | 0.3156 | 1000 | 0.9018          | 0.6730 |
| 1.0379        | 0.3787 | 1200 | 0.8131          | 0.5803 |
| 1.0379        | 0.4418 | 1400 | 0.7567          | 0.5544 |
| 0.8008        | 0.5049 | 1600 | 0.7040          | 0.5138 |
| 0.8008        | 0.5680 | 1800 | 0.6949          | 0.5236 |
| 0.7212        | 0.6311 | 2000 | 0.6722          | 0.4993 |
| 0.7212        | 0.6942 | 2200 | 0.6404          | 0.4762 |
| 0.7212        | 0.7573 | 2400 | 0.6336          | 0.4686 |
| 0.6639        | 0.8204 | 2600 | 0.5933          | 0.4587 |
| 0.6639        | 0.8836 | 2800 | 0.5996          | 0.4544 |
| 0.6278        | 0.9467 | 3000 | 0.5639          | 0.4242 |
| 0.6278        | 1.0098 | 3200 | 0.5665          | 0.4228 |
| 0.6278        | 1.0729 | 3400 | 0.5476          | 0.4191 |
| 0.5528        | 1.1360 | 3600 | 0.5440          | 0.4189 |
| 0.5528        | 1.1991 | 3800 | 0.5298          | 0.4071 |
| 0.5103        | 1.2622 | 4000 | 0.5384          | 0.4025 |
| 0.5103        | 1.3253 | 4200 | 0.5311          | 0.3998 |
| 0.5103        | 1.3885 | 4400 | 0.5396          | 0.4039 |
| 0.5194        | 1.4516 | 4600 | 0.5502          | 0.4049 |
| 0.5194        | 1.5147 | 4800 | 0.6632          | 0.4365 |
| 0.6034        | 1.5778 | 5000 | 0.7075          | 0.4368 |
| 0.6034        | 1.6409 | 5200 | 0.7466          | 0.4419 |
| 0.6034        | 1.7040 | 5400 | 0.7625          | 0.4498 |
| 0.74          | 1.7671 | 5600 | 0.7502          | 0.4475 |
| 0.74          | 1.8302 | 5800 | 0.7740          | 0.4596 |
| 0.7844        | 1.8933 | 6000 | 0.7821          | 0.4576 |


### Framework versions

- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1