metadata
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-base-google-fleurs-pt-br
results: []
whisper-base-google-fleurs-pt-br
This model is a fine-tuned version of openai/whisper-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4270
- Wer: 22.0013
- Wer Normalized: 18.1723
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: 3.05e-05
- train_batch_size: 32
- 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: 80
- training_steps: 800
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Wer Normalized |
|---|---|---|---|---|---|
| 0.6738 | 0.5 | 100 | 0.3943 | 21.7334 | 17.9487 |
| 0.4816 | 1.01 | 200 | 0.3762 | 20.9203 | 17.1352 |
| 0.2652 | 1.51 | 300 | 0.3872 | 21.1882 | 17.2827 |
| 0.2901 | 2.01 | 400 | 0.3912 | 21.4608 | 17.7061 |
| 0.1408 | 2.51 | 500 | 0.4063 | 21.6112 | 18.0010 |
| 0.1428 | 3.02 | 600 | 0.4132 | 21.8650 | 18.0201 |
| 0.0839 | 3.52 | 700 | 0.4252 | 22.3679 | 18.4720 |
| 0.0906 | 4.02 | 800 | 0.4270 | 22.0013 | 18.1723 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.1
- Datasets 2.16.1
- Tokenizers 0.15.0