whisper-large-v3-turbo-Tamil-Version1

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: 0.4012
• Wer: 59.9036

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-06
• 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
0.4551	5.8309	2000	0.4644	62.9588
0.4214	11.6618	4000	0.4396	61.9028
0.3916	17.4927	6000	0.4265	61.6157
0.3976	23.3236	8000	0.4183	61.0621
0.4002	29.1545	10000	0.4120	60.3650
0.3966	34.9854	12000	0.4076	60.0369
0.3779	40.8163	14000	0.4046	59.9754
0.3717	46.6472	16000	0.4024	59.9959
0.3763	52.4781	18000	0.4016	59.7601
0.3766	58.3090	20000	0.4012	59.9036

Framework versions

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