golesheed/whisper-non-native-children-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: 0.4064
• Wer: 11.6270

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
0.6676	0.71	30	0.4010	19.5433
0.2705	1.43	60	0.3582	13.6441
0.1694	2.14	90	0.3648	11.8934
0.0793	2.86	120	0.3757	13.0542
0.0416	3.57	150	0.3965	13.3587
0.0245	4.29	180	0.3938	11.9125
0.012	5.0	210	0.4064	11.6270

Framework versions

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