in_car_commands_26_mdl__base_ver1

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

• Loss: 3.0221
• Cer: 29.0323

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: 3.75e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
No log	1.0	1	3.0384	29.0323
No log	2.0	2	3.0384	29.0323
No log	3.0	3	3.0384	29.0323
No log	4.0	4	3.0384	29.0323
No log	5.0	5	3.0373	29.0323
No log	6.0	6	3.0384	29.0323
No log	7.0	7	3.0362	29.0323
No log	8.0	8	3.0329	29.0323
No log	9.0	9	3.0262	29.0323
No log	10.0	10	3.0221	29.0323

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu118
• Datasets 3.2.0
• Tokenizers 0.21.0