einargizz/whisper_samromur

Whisper tiny Icelandic

This model is a fine-tuned version of openai/whisper-tiny on the language-and-voice-lab/samromur_asr dataset. It achieves the following results on the evaluation set:

• Loss: 0.8061
• Wer: 53.125

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• 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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4531	0.25	250	1.5444	88.0
0.8932	0.5	500	1.0078	62.0
0.7225	0.75	750	0.8466	53.625
0.6828	1.0	1000	0.8061	53.125

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1
• Datasets 2.10.0
• Tokenizers 0.13.2