metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: whisper-small-finetuned-gtzan
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: GTZAN
type: marsyas/gtzan
config: all
split: train
args: all
metrics:
- name: Accuracy
type: accuracy
value: 0.91
whisper-small-finetuned-gtzan
This model is a fine-tuned version of openai/whisper-small on the gtzan dataset. It achieves the following results on the evaluation set:
- Loss: 0.4896
- Accuracy: 0.91
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.292 | 1.0 | 100 | 1.3642 | 0.595 |
| 1.0263 | 2.0 | 200 | 0.9241 | 0.725 |
| 0.5906 | 3.0 | 300 | 0.9602 | 0.68 |
| 0.2665 | 4.0 | 400 | 0.8529 | 0.745 |
| 0.2222 | 5.0 | 500 | 0.6671 | 0.835 |
| 0.1649 | 6.0 | 600 | 0.4792 | 0.9 |
| 0.0018 | 7.0 | 700 | 0.7901 | 0.87 |
| 0.0303 | 8.0 | 800 | 0.4475 | 0.925 |
| 0.0011 | 9.0 | 900 | 0.5972 | 0.895 |
| 0.0008 | 10.0 | 1000 | 0.5501 | 0.9 |
| 0.0007 | 11.0 | 1100 | 0.5916 | 0.895 |
| 0.0007 | 12.0 | 1200 | 0.5719 | 0.9 |
| 0.0007 | 13.0 | 1300 | 0.5082 | 0.92 |
| 0.0007 | 14.0 | 1400 | 0.4954 | 0.905 |
| 0.0006 | 15.0 | 1500 | 0.4896 | 0.91 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1