kpriyanshu256/whisper-large-v2-ga-IE-1000-32-1e-05-cy-multi

whisper-large-v2-Irish

This model is a fine-tuned version of kpriyanshu256/whisper-large-v2-cy-500-32-1e-05 on the Common Voice 11.0 and the FLEURS datasets. It achieves the following results on the evaluation set:

• Loss: 0.7879
• Wer: 34.2625

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: 8
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3768	0.25	250	0.7143	39.4890
0.1498	1.15	500	0.7663	35.8014
0.0907	2.05	750	0.7730	35.2497
0.045	2.3	1000	0.7879	34.2625

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu117
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2