indic-whisper-medium-hi-reverse-mft-1

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

• Loss: 0.1279
• Wer: 17.5313
• Cer: 7.8607

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: 0.0002
• 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: cosine
• lr_scheduler_warmup_steps: 963
• training_steps: 9621

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.1038	0.9998	2405	0.1063	18.4707	8.1700
0.0693	1.9990	4810	0.1036	17.8065	7.9757
0.0363	2.9981	7215	0.1169	17.4535	7.7419
0.029	3.9973	9620	0.1279	17.5313	7.8607

Framework versions

• Transformers 4.51.1
• Pytorch 2.8.0+cu128
• Datasets 4.4.1
• Tokenizers 0.21.4