metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: facebook/wav2vec2-xls-r-300m
metrics:
- wer
model-index:
- name: xlsr-wav2vec2-model
results: []
xlsr-wav2vec2-model
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8086
- Wer: 0.4364
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: 0.0001
- train_batch_size: 8
- 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: 1000
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 7.4037 | 2.1645 | 500 | 3.5799 | 1.0 |
| 2.9066 | 4.3290 | 1000 | 3.0843 | 1.0 |
| 2.1161 | 6.4935 | 1500 | 1.1052 | 0.7927 |
| 1.0022 | 8.6580 | 2000 | 0.7942 | 0.6240 |
| 0.6621 | 10.8225 | 2500 | 0.7538 | 0.5438 |
| 0.5084 | 12.9870 | 3000 | 0.6946 | 0.5070 |
| 0.3937 | 15.1515 | 3500 | 0.7098 | 0.4751 |
| 0.3267 | 17.3160 | 4000 | 0.7231 | 0.4620 |
| 0.2675 | 19.4805 | 4500 | 0.7481 | 0.4604 |
| 0.2308 | 21.6450 | 5000 | 0.7755 | 0.4594 |
| 0.199 | 23.8095 | 5500 | 0.7615 | 0.4438 |
| 0.1803 | 25.9740 | 6000 | 0.8217 | 0.4358 |
| 0.1599 | 28.1385 | 6500 | 0.8086 | 0.4364 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1