breco/whisper-tiny-few-audios

Whisper Tiny Few Audios - vfranchis

This model is a fine-tuned version of openai/whisper-tiny on the Few audios 1.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3835
• Wer: 15.7143

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• training_steps: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.5625	2.8571	10	1.4533	77.1429
0.6893	5.7143	20	0.7903	32.8571
0.1921	8.5714	30	0.5135	34.2857
0.0623	11.4286	40	0.4158	11.4286
0.0222	14.2857	50	0.3903	14.2857
0.0107	17.1429	60	0.3846	14.2857
0.0069	20.0	70	0.3847	15.7143
0.0055	22.8571	80	0.3842	15.7143
0.0046	25.7143	90	0.3836	15.7143
0.0044	28.5714	100	0.3835	15.7143

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1