metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-small-CV_Fleurs_AMMI_ALFFA-sw-100hrs-v1
results: []
whisper-small-CV_Fleurs_AMMI_ALFFA-sw-100hrs-v1
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4908
- Wer: 0.1975
- Cer: 0.0747
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use adamw_hf with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.766 | 1.0 | 3941 | 0.4901 | 0.3532 | 0.1359 |
| 0.5754 | 2.0 | 7882 | 0.3695 | 0.2248 | 0.0827 |
| 0.3572 | 3.0 | 11823 | 0.3460 | 0.1895 | 0.0662 |
| 0.2438 | 4.0 | 15764 | 0.3584 | 0.2005 | 0.0812 |
| 0.1872 | 5.0 | 19705 | 0.3732 | 0.2092 | 0.0885 |
| 0.1594 | 6.0 | 23646 | 0.3998 | 0.1917 | 0.0704 |
| 0.1483 | 7.0 | 27587 | 0.4195 | 0.2008 | 0.0754 |
| 0.1437 | 8.0 | 31528 | 0.4354 | 0.2083 | 0.0791 |
| 0.1413 | 9.0 | 35469 | 0.4415 | 0.1969 | 0.0729 |
| 0.1392 | 10.0 | 39410 | 0.4553 | 0.1995 | 0.0766 |
| 0.1312 | 11.0 | 43351 | 0.4681 | 0.2010 | 0.0786 |
| 0.1095 | 12.0 | 47292 | 0.4726 | 0.2014 | 0.0819 |
| 0.0945 | 13.0 | 51233 | 0.4908 | 0.1975 | 0.0747 |
Framework versions
- Transformers 4.46.1
- Pytorch 2.1.0+cu118
- Datasets 3.1.0
- Tokenizers 0.20.1