metadata
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-base-khmer-aug
results: []
whisper-base-khmer-aug
This model is a fine-tuned version of openai/whisper-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2505
- Wer: 63.2723
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 |
---|---|---|---|---|
0.6835 | 1.0 | 670 | 0.2957 | 79.4714 |
0.2814 | 2.0 | 1340 | 0.2385 | 73.0988 |
0.2102 | 3.0 | 2010 | 0.2247 | 68.2828 |
0.1726 | 4.0 | 2680 | 0.2124 | 67.2288 |
0.147 | 5.0 | 3350 | 0.2230 | 66.5802 |
0.1243 | 6.0 | 4020 | 0.2268 | 72.6123 |
0.1078 | 7.0 | 4690 | 0.2268 | 65.4775 |
0.0949 | 8.0 | 5360 | 0.2329 | 63.2074 |
0.0834 | 9.0 | 6030 | 0.2507 | 63.2074 |
0.0732 | 10.0 | 6700 | 0.2505 | 63.2723 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.3.1
- Datasets 2.21.0
- Tokenizers 0.19.1