ntnu-smil/whisper-large-v3-sandi-7k-1024-28steps-merged

whisper-large-v3-sandi-7k-1024-28steps

This model is a fine-tuned version of openai/whisper-large-v3 on the ntnu-smil/sandi2025-ds dataset. It achieves the following results on the evaluation set:

• Loss: 1.0275
• Wer: 73.8237
• Cer: 203.1654
• Decode Runtime: 257.7123
• Wer Runtime: 0.2072
• Cer Runtime: 0.4935

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: 7e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 1024
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 28

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer	Decode Runtime	Wer Runtime	Cer Runtime
1.9026	1.0357	7	1.3670	70.5577	206.9010	266.1791	0.2147	0.5053
1.2477	2.0714	14	1.1783	86.2572	223.6346	268.9910	0.2241	0.5015
1.07	3.1071	21	1.0605	78.7713	211.1141	262.5822	0.2186	0.5076
1.0348	4.1429	28	1.0275	73.8237	203.1654	257.7123	0.2072	0.4935

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.4.1+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1