amanjain96/whisper-small-hi

Whisper Small Hi - Oriya

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.5345
• Wer: 58.7293

Model description

This model is fine tuned for the language oriya to support ASR.

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-05
• train_batch_size: 16
• 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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0042	20.0	1000	0.3678	61.9920
0.0013	40.0	2000	0.4025	59.0155
0.0	60.0	3000	0.4559	56.8975
0.0	80.0	4000	0.5129	58.5575
0.0	100.0	5000	0.5345	58.7293

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1