multilingual-whisper-v3-turbo-nchlt

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2263
• Wer: 0.1647
• Cer: 0.0453

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.3933	1.0215	500	0.3577	0.2719	0.0586
0.1124	2.043	1000	0.2711	0.1921	0.0402
0.0602	3.0645	1500	0.2412	0.1727	0.0387
0.026	4.086	2000	0.2263	0.1647	0.0453

Framework versions

• Transformers 4.52.0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.4