metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
datasets:
- swagen
metrics:
- wer
model-index:
- name: whisper-medium-swagen-male-model-test
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: swagen
type: swagen
metrics:
- name: Wer
type: wer
value: 0.3229230455194938
whisper-medium-swagen-male-model-test
This model is a fine-tuned version of openai/whisper-medium on the swagen dataset. It achieves the following results on the evaluation set:
- Loss: 0.5615
- Wer: 0.3229
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.7031 | 0.4840 | 200 | 0.7854 | 0.5021 |
| 2.0469 | 0.9679 | 400 | 0.6388 | 0.3989 |
| 1.0941 | 1.4501 | 600 | 0.5956 | 0.3358 |
| 1.0543 | 1.9341 | 800 | 0.5615 | 0.3229 |
| 0.4611 | 2.4162 | 1000 | 0.5803 | 0.3070 |
| 0.5304 | 2.9002 | 1200 | 0.5714 | 0.3460 |
| 0.2124 | 3.3823 | 1400 | 0.6000 | 0.2872 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0