wwwtwwwt/whisper-tiny-Education-Documentaries-testing

Whisper Tiny En - Education - Documentaries

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

• Loss: 1.2666
• Wer: 50.2757

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: Use adamw_torch 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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8091	0.8230	1000	1.3021	62.2507
0.5487	1.6461	2000	1.2534	58.9165
0.4782	2.4691	3000	1.2633	53.8332
0.3527	3.2922	4000	1.2666	50.2757

Framework versions

• Transformers 4.49.0
• Pytorch 2.4.0
• Datasets 3.3.2
• Tokenizers 0.21.0