whisper-large-v3-genbed-f

This model is a fine-tuned version of openai/whisper-large-v3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4613
• Wer: 28.2294

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: 1.75e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 2500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.4575	0.6605	250	0.5118	48.6061
0.3575	1.3210	500	0.4580	41.5408
0.3229	1.9815	750	0.3920	34.9542
0.1937	2.6420	1000	0.4103	33.1986
0.0955	3.3025	1250	0.4218	32.8368
0.0943	3.9630	1500	0.4120	31.6982
0.0346	4.6235	1750	0.4397	30.2724
0.0123	5.2840	2000	0.4604	28.8891
0.0132	5.9445	2250	0.4485	29.1658
0.0025	6.6050	2500	0.4613	28.2294

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1