golesheed/whisper-native-elderly-4-dutch

Whisper Large V2

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3247
• Wer: 13.4709

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 20
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5388	0.49	30	0.3297	12.2434
0.2858	0.98	60	0.2893	23.3419
0.143	1.48	90	0.2922	13.5327
0.1337	1.97	120	0.2838	10.7065
0.0606	2.46	150	0.2905	10.3765
0.0557	2.95	180	0.2915	10.0258
0.0265	3.44	210	0.3139	10.8613
0.0207	3.93	240	0.3094	10.0670
0.0098	4.43	270	0.3188	12.0578
0.0098	4.92	300	0.3247	13.4709

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.14.6
• Tokenizers 0.15.0