whisper_tn_hi_v3

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

• Loss: 0.5846
• Wer: 78.6633
• Cer: 293.0339

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1.6414	1.0	409	0.8810	83.6367	247.9323
0.5322	2.0	818	0.6398	76.073	285.8609
0.3633	3.0	1227	0.5905	70.2743	268.7294
0.2979	4.0	1636	0.5846	78.6633	293.0339

Framework versions

• Transformers 4.24.0
• Pytorch 1.13.0
• Datasets 2.6.1
• Tokenizers 0.11.0