metadata
library_name: transformers
license: apache-2.0
base_model: elgeish/wav2vec2-large-xlsr-53-arabic
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: result
results: []
result
This model is a fine-tuned version of elgeish/wav2vec2-large-xlsr-53-arabic on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5629
- Wer: 0.5924
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: 0.001
- train_batch_size: 4
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 8.2289 | 2.0 | 500 | 1.8545 | 0.9936 |
| 1.2421 | 4.0 | 1000 | 1.5310 | 0.9443 |
| 1.0752 | 6.0 | 1500 | 1.3956 | 0.9045 |
| 0.6889 | 8.0 | 2000 | 0.7714 | 0.6768 |
| 0.3904 | 10.0 | 2500 | 0.5629 | 0.5924 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0
- Datasets 3.1.0
- Tokenizers 0.20.3