|
--- |
|
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: 0.4589 |
|
- Wer: 0.3743 |
|
|
|
## 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: 2500 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:------:|:----:|:---------------:|:------:| |
|
| No log | 0.1414 | 100 | 4.0354 | 1.0 | |
|
| No log | 0.2829 | 200 | 3.0977 | 1.0 | |
|
| No log | 0.4243 | 300 | 3.0769 | 1.0 | |
|
| No log | 0.5658 | 400 | 1.3738 | 0.8914 | |
|
| 3.7586 | 0.7072 | 500 | 1.0915 | 0.7692 | |
|
| 3.7586 | 0.8487 | 600 | 0.9361 | 0.6855 | |
|
| 3.7586 | 0.9901 | 700 | 0.8495 | 0.6247 | |
|
| 3.7586 | 1.1315 | 800 | 0.6886 | 0.5397 | |
|
| 3.7586 | 1.2730 | 900 | 0.6704 | 0.5312 | |
|
| 0.8877 | 1.4144 | 1000 | 0.6237 | 0.4950 | |
|
| 0.8877 | 1.5559 | 1100 | 0.5992 | 0.4768 | |
|
| 0.8877 | 1.6973 | 1200 | 0.5730 | 0.4522 | |
|
| 0.8877 | 1.8388 | 1300 | 0.5504 | 0.4418 | |
|
| 0.8877 | 1.9802 | 1400 | 0.5288 | 0.4259 | |
|
| 0.6844 | 2.1216 | 1500 | 0.5165 | 0.4217 | |
|
| 0.6844 | 2.2631 | 1600 | 0.5072 | 0.4193 | |
|
| 0.6844 | 2.4045 | 1700 | 0.4984 | 0.4155 | |
|
| 0.6844 | 2.5460 | 1800 | 0.4882 | 0.4097 | |
|
| 0.6844 | 2.6874 | 1900 | 0.4804 | 0.4080 | |
|
| 0.537 | 2.8289 | 2000 | 0.4700 | 0.3927 | |
|
| 0.537 | 2.9703 | 2100 | 0.4677 | 0.3885 | |
|
| 0.537 | 3.1117 | 2200 | 0.4683 | 0.3857 | |
|
| 0.537 | 3.2532 | 2300 | 0.4618 | 0.3792 | |
|
| 0.537 | 3.3946 | 2400 | 0.4604 | 0.3763 | |
|
| 0.4434 | 3.5361 | 2500 | 0.4589 | 0.3743 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
