finetune_v9

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

• Loss: 0.4229
• Wer: 110.4363

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: 16
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• training_steps: 80
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	10.0	10	0.5239	19.2472
No log	20.0	20	0.4348	17.8785
No log	30.0	30	0.4055	17.2797
No log	40.0	40	0.4204	18.5629
0.0997	50.0	50	0.4292	20.7015
0.0997	60.0	60	0.4282	19.6749
0.0997	70.0	70	0.4246	47.5620
0.0997	80.0	80	0.4229	110.4363

Framework versions

• Transformers 4.42.3
• Pytorch 2.2.0
• Datasets 2.20.0
• Tokenizers 0.19.1