Jaspernl/whisper-tiny-ft-nl_ortho

Whisper Tiny FT NL_bs32

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 13.0, voxpopuli dataset. It achieves the following results on the evaluation set:

• Loss: 0.2896
• Wer: 97.7034

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: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 10000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3004	0.1	1000	0.3763	82.0359
0.1669	1.07	2000	0.3136	53.1062
0.1276	2.03	3000	0.2915	64.4126
0.1146	2.13	4000	0.3259	60.0555
0.0758	3.1	5000	0.3042	73.8444
0.059	4.06	6000	0.3088	83.9768
0.0554	5.03	7000	0.2905	81.0178
0.0503	5.13	8000	0.2877	88.2464
0.0499	6.1	9000	0.3116	91.4979
0.0493	7.06	10000	0.2896	97.7034

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.1
• Datasets 2.14.7
• Tokenizers 0.14.1