---
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-base-khmer-aug
  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. -->

# whisper-base-khmer-aug

This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2816
- Wer: 67.1639

## 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: 16
- eval_batch_size: 16
- 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: constant
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 1.0104        | 1.0   | 670  | 0.4883          | 90.8059 |
| 0.5473        | 2.0   | 1340 | 0.3166          | 92.7680 |
| 0.4257        | 3.0   | 2010 | 0.2780          | 78.1904 |
| 0.364         | 4.0   | 2680 | 0.2614          | 78.2552 |
| 0.3197        | 5.0   | 3350 | 0.2493          | 71.9150 |
| 0.2863        | 6.0   | 4020 | 0.2539          | 73.3420 |
| 0.2574        | 7.0   | 4690 | 0.2557          | 69.5962 |
| 0.2358        | 8.0   | 5360 | 0.2655          | 69.6935 |
| 0.2154        | 9.0   | 6030 | 0.2662          | 68.7693 |
| 0.1956        | 10.0  | 6700 | 0.2816          | 67.1639 |


### Framework versions

- Transformers 4.44.0
- Pytorch 2.3.1
- Datasets 2.21.0
- Tokenizers 0.19.1