whisper-kaggle-be3

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

• Loss: 0.3213
• Wer: 45.6094

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0749	3.15	1000	0.1886	50.9273
0.0117	6.31	2000	0.2440	47.2937
0.0014	9.46	3000	0.2978	45.8933
0.0003	12.62	4000	0.3213	45.6094

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.0
• Datasets 2.14.4.dev0
• Tokenizers 0.13.3