whisper-large-v3-turbo-Hindi-Version2

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.2175
• Wer: 23.1550

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use 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: 1000
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2178	6.7797	2000	0.2245	25.6931
0.1841	13.5593	4000	0.2146	24.6095
0.1572	20.3390	6000	0.2121	23.5845
0.1489	27.1186	8000	0.2120	23.9848
0.1315	33.8983	10000	0.2118	23.6822
0.1253	40.6780	12000	0.2145	22.9793
0.1154	47.4576	14000	0.2154	23.1941
0.1151	54.2373	16000	0.2168	23.0964
0.1079	61.0169	18000	0.2175	23.1550

Framework versions

• PEFT 0.14.0
• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.20.1