kul-speech-lab/whisper-small-nl-dy

Whisper Small Dutch

This model is a fine-tuned version of qmeeus/whisper-small-nl on the data/copas copas-full dataset. It achieves the following results on the evaluation set:

• Loss: 0.0015
• Wer: 0.0

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1291	2.03	500	0.2953	25.0681
0.0524	5.03	1000	0.1626	13.9118
0.0403	8.03	1500	0.0825	6.5450
0.0349	11.03	2000	0.0409	2.5652
0.0122	14.03	2500	0.0173	0.6619
0.0053	17.03	3000	0.0068	0.0822
0.0032	20.02	3500	0.0037	0.0173
0.0022	23.02	4000	0.0024	0.0
0.0018	26.02	4500	0.0018	0.0
0.0016	29.02	5000	0.0015	0.0

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.12.1+cu116
• Datasets 2.4.0
• Tokenizers 0.12.1