whisper-large-v3-finetuned-7

This model is a fine-tuned version of openai/whisper-large-v3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2896
• Wer: 19.3061

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: 2e-09
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.9515	1.0	7532	1.7699	27.1166
1.2475	2.0	15064	1.6434	24.6623
1.2258	3.0	22596	1.5462	22.7052
0.4176	4.0	30128	1.4713	21.4012
1.2809	5.0	37660	1.4125	20.3966
1.5466	6.0	45192	1.3656	19.7720
0.6952	7.0	52724	1.3283	19.4675
2.0103	8.0	60256	1.3030	19.3087
1.4738	9.0	67788	1.2914	19.3087
3.545	10.0	75320	1.2896	19.3061

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2