metadata
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: wav2vec2-large-xlsr-53-urdu
results: []
wav2vec2-large-xlsr-53-urdu
This model is a fine-tuned version of Harveenchadha/vakyansh-wav2vec2-urdu-urm-60 on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 9.8476
- Wer: 0.6253
- Cer: 0.3456
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: 200
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 13.7919 | 8.33 | 100 | 7.9423 | 0.6742 | 0.3737 |
| 1.4283 | 16.67 | 200 | 3.2921 | 0.6681 | 0.3707 |
| 0.4868 | 25.0 | 300 | 9.8476 | 0.6253 | 0.3456 |
Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3