golesheed/whisper-native-elderly-8-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: 0.2110
• Wer: 7.8855

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
0.5451	0.49	30	0.2331	9.3771
0.299	0.98	60	0.1932	11.1867
0.1459	1.48	90	0.1867	11.2415
0.1368	1.97	120	0.1862	10.0022
0.0654	2.46	150	0.1957	13.2156
0.0592	2.95	180	0.1975	9.3222
0.031	3.44	210	0.2102	8.1377
0.0226	3.93	240	0.1986	7.8965
0.0119	4.43	270	0.2104	8.0061
0.0094	4.92	300	0.2110	7.8855

Framework versions

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