metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: >-
subhadeep_whisper_small_finetune_teacher_no_noise_libri_100_hours_100_epochs_batch_8
results: []
subhadeep_whisper_small_finetune_teacher_no_noise_libri_100_hours_100_epochs_batch_8
This model is a fine-tuned version of openai/whisper-small.en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3422
- Wer: 15.6961
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: 0.0005
- train_batch_size: 4
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 256
- total_train_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3854 | 3.68 | 100 | 0.1916 | 12.3782 |
| 0.0214 | 7.39 | 200 | 0.2210 | 12.6356 |
| 0.0209 | 11.11 | 300 | 0.2540 | 13.5455 |
| 0.0698 | 14.79 | 400 | 0.2788 | 13.9829 |
| 0.0206 | 18.5 | 500 | 0.3106 | 14.8156 |
| 0.0236 | 22.22 | 600 | 0.3422 | 15.6961 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.8.0
- Tokenizers 0.13.2