whisper-large-v3-Urdu-Version1

This model is a fine-tuned version of openai/whisper-large-v3 on the fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.3244
• Wer: 20.6725

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-06
• train_batch_size: 8
• 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: 1000
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3271	6.7340	2000	0.3375	21.5842
0.3107	13.4680	4000	0.3244	20.9093
0.2797	20.2020	6000	0.3205	20.8383
0.2639	26.9360	8000	0.3202	20.5778
0.2529	33.6700	10000	0.3216	20.7909
0.26	40.4040	12000	0.3230	20.6843
0.2485	47.1380	14000	0.3244	20.6725

Framework versions

• PEFT 0.12.1.dev0
• Transformers 4.45.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1