|
--- |
|
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: 0.4105 |
|
- Wer: 0.3136 |
|
|
|
## 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 | 3.2195 | 1.0 | |
|
| No log | 0.2829 | 200 | 3.1180 | 1.0 | |
|
| No log | 0.4243 | 300 | 1.7236 | 0.9364 | |
|
| No log | 0.5658 | 400 | 1.1579 | 0.7628 | |
|
| 3.0279 | 0.7072 | 500 | 0.9558 | 0.6721 | |
|
| 3.0279 | 0.8487 | 600 | 0.8189 | 0.6424 | |
|
| 3.0279 | 0.9901 | 700 | 0.6973 | 0.5164 | |
|
| 3.0279 | 1.1315 | 800 | 0.6183 | 0.4752 | |
|
| 3.0279 | 1.2730 | 900 | 0.5936 | 0.4703 | |
|
| 0.7925 | 1.4144 | 1000 | 0.5498 | 0.4304 | |
|
| 0.7925 | 1.5559 | 1100 | 0.5286 | 0.4148 | |
|
| 0.7925 | 1.6973 | 1200 | 0.5130 | 0.3974 | |
|
| 0.7925 | 1.8388 | 1300 | 0.4878 | 0.3864 | |
|
| 0.7925 | 1.9802 | 1400 | 0.4740 | 0.3733 | |
|
| 0.62 | 2.1216 | 1500 | 0.4578 | 0.3593 | |
|
| 0.62 | 2.2631 | 1600 | 0.4535 | 0.3500 | |
|
| 0.62 | 2.4045 | 1700 | 0.4485 | 0.3497 | |
|
| 0.62 | 2.5460 | 1800 | 0.4373 | 0.3431 | |
|
| 0.62 | 2.6874 | 1900 | 0.4362 | 0.3429 | |
|
| 0.4879 | 2.8289 | 2000 | 0.4236 | 0.3327 | |
|
| 0.4879 | 2.9703 | 2100 | 0.4172 | 0.3257 | |
|
| 0.4879 | 3.1117 | 2200 | 0.4206 | 0.3217 | |
|
| 0.4879 | 3.2532 | 2300 | 0.4166 | 0.3199 | |
|
| 0.4879 | 3.3946 | 2400 | 0.4134 | 0.3173 | |
|
| 0.4036 | 3.5361 | 2500 | 0.4110 | 0.3159 | |
|
| 0.4036 | 3.6775 | 2600 | 0.4105 | 0.3136 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
