metadata
library_name: transformers
language:
- en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: bhattasp/whisper-finetuned-all-3
results: []
bhattasp/whisper-finetuned-all-3
This model is a fine-tuned version of openai/whisper-tiny on the AtCO2_UWB, Atcosim, Bial dataset. It achieves the following results on the evaluation set:
- Loss: 0.3387
- Wer: 22.9038
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4559 | 1.0 | 1342 | 0.4642 | 36.8026 |
| 0.2906 | 2.0 | 2684 | 0.3681 | 24.1868 |
| 0.1993 | 3.0 | 4026 | 0.3436 | 22.6871 |
| 0.1389 | 4.0 | 5368 | 0.3372 | 22.8121 |
| 0.1308 | 5.0 | 6710 | 0.3387 | 22.9038 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0