metadata
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: []
wav2vec2-xlsr-53-ft-btb-ccv-cy
This model is a fine-tuned version of 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: 0.4200
- Wer: 0.3227
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: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 0.1548 | 100 | 3.5623 | 1.0 |
| No log | 0.3096 | 200 | 3.2967 | 1.0 |
| No log | 0.4644 | 300 | 2.6484 | 1.0000 |
| No log | 0.6192 | 400 | 1.0602 | 0.7315 |
| 3.6398 | 0.7740 | 500 | 0.8942 | 0.6696 |
| 3.6398 | 0.9288 | 600 | 0.7116 | 0.5361 |
| 3.6398 | 1.0836 | 700 | 0.6648 | 0.5101 |
| 3.6398 | 1.2384 | 800 | 0.5869 | 0.4528 |
| 3.6398 | 1.3932 | 900 | 0.5698 | 0.4359 |
| 0.5976 | 1.5480 | 1000 | 0.5408 | 0.4112 |
| 0.5976 | 1.7028 | 1100 | 0.5229 | 0.4196 |
| 0.5976 | 1.8576 | 1200 | 0.5055 | 0.3955 |
| 0.5976 | 2.0124 | 1300 | 0.4808 | 0.3709 |
| 0.5976 | 2.1672 | 1400 | 0.4667 | 0.3580 |
| 0.443 | 2.3220 | 1500 | 0.4573 | 0.3582 |
| 0.443 | 2.4768 | 1600 | 0.4475 | 0.3452 |
| 0.443 | 2.6316 | 1700 | 0.4369 | 0.3478 |
| 0.443 | 2.7864 | 1800 | 0.4227 | 0.3298 |
| 0.443 | 2.9412 | 1900 | 0.4169 | 0.3271 |
| 0.3475 | 3.0960 | 2000 | 0.4200 | 0.3227 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1