outout_model_shunya_100k_steps

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

• Loss: 0.8833
• Wer: 32.8460

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 4
• seed: 42
• optimizer: Use OptimizerNames.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: 200
• training_steps: 100000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1042	7.8125	10000	0.8219	44.7054
0.0466	15.625	20000	0.8904	41.6334
0.0220	23.4375	30000	0.8833	39.8169
0.0226	31.25	40000	0.9156	40.6735
0.0069	39.0625	50000	0.9137	36.9812
0.0027	46.875	60000	0.9308	36.9517
0.0016	54.6875	70000	0.9383	34.9136
0.0005	62.5	80000	0.9123	34.6478
0.0000	70.3125	90000	0.8958	33.3629
0.0000	78.125	100000	0.8833	32.8460

Framework versions

• Transformers 5.0.1.dev0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2