whisper-small-ru-v3

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

• Loss: 0.1974
• Wer: 16.5586

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: 32
• eval_batch_size: 16
• 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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2389	0.2413	200	0.2515	22.1606
0.2039	0.4825	400	0.2257	18.2687
0.1946	0.7238	600	0.2135	17.4667
0.1775	0.9650	800	0.2001	16.7119
0.1037	1.2063	1000	0.1974	16.5586

Framework versions

• Transformers 4.49.0
• Pytorch 2.1.0+cu118
• Datasets 3.3.1
• Tokenizers 0.21.0