---
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: 0.4593
- Wer: 0.3553

## 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: 64
- eval_batch_size: 8
- seed: 42
- 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        | 0.0566 | 100  | 3.5994          | 1.0    |
| No log        | 0.1133 | 200  | 3.0224          | 1.0    |
| No log        | 0.1699 | 300  | 1.9627          | 0.9061 |
| No log        | 0.2265 | 400  | 0.9921          | 0.6952 |
| 3.6184        | 0.2831 | 500  | 0.8504          | 0.6158 |
| 3.6184        | 0.3398 | 600  | 0.7936          | 0.5964 |
| 3.6184        | 0.3964 | 700  | 0.7601          | 0.5606 |
| 3.6184        | 0.4530 | 800  | 0.6909          | 0.5128 |
| 3.6184        | 0.5096 | 900  | 0.6617          | 0.4916 |
| 0.4662        | 0.5663 | 1000 | 0.6467          | 0.4812 |
| 0.4662        | 0.6229 | 1100 | 0.6144          | 0.4639 |
| 0.4662        | 0.6795 | 1200 | 0.5942          | 0.4537 |
| 0.4662        | 0.7361 | 1300 | 0.5675          | 0.4351 |
| 0.4662        | 0.7928 | 1400 | 0.5539          | 0.4230 |
| 0.3508        | 0.8494 | 1500 | 0.5448          | 0.4144 |
| 0.3508        | 0.9060 | 1600 | 0.5326          | 0.4066 |
| 0.3508        | 0.9626 | 1700 | 0.5155          | 0.3989 |
| 0.3508        | 1.0193 | 1800 | 0.5067          | 0.3856 |
| 0.3508        | 1.0759 | 1900 | 0.4916          | 0.3724 |
| 0.2774        | 1.1325 | 2000 | 0.4855          | 0.3697 |
| 0.2774        | 1.1891 | 2100 | 0.4798          | 0.3661 |
| 0.2774        | 1.2458 | 2200 | 0.4774          | 0.3646 |
| 0.2774        | 1.3024 | 2300 | 0.4699          | 0.3585 |
| 0.2774        | 1.3590 | 2400 | 0.4651          | 0.3550 |
| 0.2328        | 1.4156 | 2500 | 0.4611          | 0.3570 |
| 0.2328        | 1.4723 | 2600 | 0.4593          | 0.3553 |


### Framework versions

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