whisper-10hrs-class-seed2025

This model is a fine-tuned version of openai/whisper-large-v2 on the JASMIN-CGN dataset. It achieves the following results on the evaluation set:

• Loss: 0.3592
• Wer: 16.7310

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: 48
• eval_batch_size: 32
• 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: 26
• num_epochs: 3.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8601	0.2955	26	0.9317	34.3074
0.4536	0.5909	52	0.4684	21.0018
0.4433	0.8864	78	0.4094	18.6869
0.382	1.1818	104	0.3878	17.5764
0.3563	1.4773	130	0.3767	17.3281
0.3275	1.7727	156	0.3711	17.2376
0.3462	2.0682	182	0.3649	17.3684
0.3265	2.3636	208	0.3617	17.0933
0.33	2.6591	234	0.3596	17.0363
0.3168	2.9545	260	0.3592	16.7310

Framework versions

• PEFT 0.16.0
• Transformers 4.57.3
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.22.2