turasa/whisper-small-ka-test

This model is a fine-tuned version of openai/whisper-small on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0471
• Wer: 30.6891

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 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: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0648	0.2466	1000	0.0841	45.7440
0.0455	0.4932	2000	0.0612	36.4233
0.0398	0.7398	3000	0.0514	32.4669
0.036	0.9864	4000	0.0471	30.6891

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0