whisper-medium-LoRA

This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3284
• Wer: 12.2024

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: 0.001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.6785	1.0	789	0.4078	18.0952
0.3283	2.0	1578	0.3660	19.9802
0.2298	3.0	2367	0.3391	14.2063
0.1458	4.0	3156	0.3307	13.7302
0.0688	5.0	3945	0.3211	12.5
0.0222	6.0	4734	0.3284	12.2024

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3