metadata
library_name: transformers
base_model: openai/whisper-large-turbo
tags:
- generated_from_trainer
datasets:
- Farhang87/whisperlargeturbo-Medical-ASR-Data
metrics:
- wer
model-index:
- name: Whisper Large Turbo Medical
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Medical ASR
type: Farhang87/whisperlargeturbo-Medical-ASR-Data
metrics:
- name: Wer
type: wer
value: 4.344677769732078
Whisper Large Turbo Medical
This model is a fine-tuned version of openai/whisper-large-turbo on the Medical ASR dataset. It achieves the following results on the evaluation set:
- Loss: 0.0672
- Wer: 4.3447
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: 32
- eval_batch_size: 8
- seed: 42
- 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: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2994 | 0.5405 | 100 | 0.2101 | 9.9928 |
| 0.1405 | 1.0811 | 200 | 0.1212 | 5.7929 |
| 0.0859 | 1.6216 | 300 | 0.0929 | 4.4895 |
| 0.044 | 2.1622 | 400 | 0.0739 | 3.9585 |
| 0.0248 | 2.7027 | 500 | 0.0672 | 4.3447 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.4.0+cu121
- Datasets 3.3.2
- Tokenizers 0.21.0