indic-whisper-kn-reverse-mft-1-0-1-5

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0819
• Wer: 26.6074
• Cer: 7.3916

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: 6.25e-06
• train_batch_size: 64
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• 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: 56
• training_steps: 560

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.098	1.0	140	0.0895	27.1272	7.7927
0.092	2.0	280	0.0836	26.8263	7.7415
0.0902	3.0	420	0.0822	26.6689	7.4245
0.0899	4.0	560	0.0819	26.6074	7.3916

Framework versions

• Transformers 4.51.1
• Pytorch 2.7.1+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4