This model is a fine-tuned version of openai/whisper-small on the Audio Emotions Collection dataset. It achieves the following results on the evaluation set:

Loss: 0.9543

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: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 5000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.672	1.0	685	0.7548
0.4456	2.0	1370	0.6081
0.2742	3.0	2055	0.5351
0.2174	4.0	2740	0.7506
0.0612	5.0	3425	1.0163
0.0661	6.0	4110	0.9165
0.0276	7.0	4795	0.9554
0.0007	7.2993	5000	0.9543

Framework versions

Transformers 4.57.3
Pytorch 2.9.0+cu126
Datasets 4.4.2
Tokenizers 0.22.1

lucasvmigotto/whisper-small-audio-emotion-classification

Whisper Small - Audio Emotion Classifier

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

lucasvmigotto/
whisper-small-audio-emotion-classification