CKSINGH/whisper-small-hi-finalBatch

Whisper small Hi - CKS 2411 finalbatch

This model is a fine-tuned version of CKSINGH/whisper-small-hi-graminVoice on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4932
• Wer: 34.6890
• Cer: 23.4230

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: 1.75e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.1262	8.77	500	0.4595	29.7847	17.7683
0.006	17.54	1000	0.4932	34.6890	23.4230

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0