---
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: wav2vec2-large-xlsr-53-urdu
  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-large-xlsr-53-urdu

This model is a fine-tuned version of [Harveenchadha/vakyansh-wav2vec2-urdu-urm-60](https://huggingface.co/Harveenchadha/vakyansh-wav2vec2-urdu-urm-60) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Loss: 11.1562
- Wer: 0.5921
- Cer: 0.3288

## 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: 50
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
| 13.83         | 8.33  | 100  | 6.3080          | 0.6611 | 0.3639 |
| 1.0144        | 16.67 | 200  | 7.7685          | 0.6498 | 0.3731 |
| 0.5801        | 25.0  | 300  | 7.8429          | 0.6454 | 0.3767 |
| 0.3344        | 33.33 | 400  | 5.9906          | 0.6349 | 0.3548 |
| 0.1606        | 41.67 | 500  | 10.8584         | 0.6105 | 0.3348 |
| 0.0974        | 50.0  | 600  | 11.1562         | 0.5921 | 0.3288 |


### Framework versions

- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3