kiranpantha/whisper-large-v3-nepali-lora-qkv

kiranpantha/whisper-large-v3-nepali

This model is a fine-tuned version of kiranpantha/whisper-large-v3-nepali on the OpenSLR54 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2859

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	9	0.5857
No log	2.0	18	0.2806
0.4997	3.0	27	0.2859

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.5.1+cxx11.abi
• Datasets 3.2.0
• Tokenizers 0.21.0