nrlt/whisper-small-mn-last

Whisper Small - Mongolian LoRA

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

• Loss: 0.2421
• Wer: 0.2158

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-4
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.3506	1.6892	500	0.3970	0.3962
0.7855	3.3784	1000	0.2958	0.3157
0.5229	5.0676	1500	0.2583	0.2628
0.2567	6.7568	2000	0.2361	0.2334
0.0620	8.4459	2500	0.2421	0.2158
0.0178	10.000	2960	0.2421	0.2158

Framework versions

• PEFT 0.18.1
• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2