metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- emodb
metrics:
- accuracy
language:
- de
model-index:
- name: whisper-tiny-de-emodb-emotion-classification
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: Emo-DB
type: emodb
metrics:
- name: Accuracy
type: accuracy
value: 0.9158878504672897
whisper-tiny-de-emodb-emotion-classification
This model is a fine-tuned version of openai/whisper-tiny on the german Emo-DB dataset. It achieves the following results on the evaluation set:
- Loss: 0.4912
- Accuracy: 0.9159
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: 2
- eval_batch_size: 2
- seed: 42
- 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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3193 | 1.0 | 214 | 1.4616 | 0.3925 |
| 0.1342 | 2.0 | 428 | 1.0384 | 0.6449 |
| 0.0582 | 3.0 | 642 | 1.5578 | 0.6542 |
| 0.6567 | 4.0 | 856 | 1.2043 | 0.7850 |
| 0.0202 | 5.0 | 1070 | 0.5967 | 0.8598 |
| 0.0008 | 6.0 | 1284 | 0.6261 | 0.8692 |
| 0.0006 | 7.0 | 1498 | 0.5857 | 0.8785 |
| 0.0004 | 8.0 | 1712 | 0.4992 | 0.9065 |
| 0.0004 | 9.0 | 1926 | 0.4943 | 0.9159 |
| 0.0003 | 10.0 | 2140 | 0.4912 | 0.9159 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.4.0+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1