amk-whisper-v5.5

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8049
• Wer: 28.0612

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: 5e-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.2935	1.0	1	1.0842	38.2653
1.2935	2.0	2	1.0339	35.7143
1.1811	3.0	3	0.9826	33.1633
0.9953	4.0	4	0.9103	28.5714
0.7994	5.0	5	0.8852	28.0612
0.7257	6.0	6	0.8624	28.5714
0.656	7.0	7	0.8423	28.8265
0.5976	8.0	8	0.8274	28.5714
0.5572	9.0	9	0.8110	27.8061
0.5096	10.0	10	0.8049	28.0612

Framework versions

• Transformers 4.29.0.dev0
• Pytorch 1.13.1+cu117
• Datasets 2.8.0
• Tokenizers 0.13.2