metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: whisper_speechcommandsV2_final
results: []
whisper_speechcommandsV2_final
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: 0.4688
- Accuracy: 0.9061
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: 3e-05
- train_batch_size: 42
- eval_batch_size: 42
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 168
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.1229 | 1.0 | 505 | 0.4320 | 0.8990 |
| 0.0833 | 2.0 | 1010 | 0.4022 | 0.9033 |
| 0.0548 | 3.0 | 1515 | 0.3930 | 0.9055 |
| 0.0566 | 4.0 | 2021 | 0.4313 | 0.9051 |
| 0.0565 | 5.0 | 2526 | 0.4355 | 0.9059 |
| 0.0165 | 6.0 | 3031 | 0.4096 | 0.9065 |
| 0.0181 | 7.0 | 3536 | 0.4436 | 0.9057 |
| 0.017 | 8.0 | 4042 | 0.4663 | 0.9061 |
| 0.0077 | 9.0 | 4547 | 0.4599 | 0.9065 |
| 0.0042 | 10.0 | 5050 | 0.4688 | 0.9061 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1