Devanshj7/whisper-english

whisper-small-hi-2

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

• Loss: 0.2909
• Wer: 7.6142

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1271	2.63	100	0.2529	7.3604
0.0246	5.26	200	0.2495	8.6294
0.0087	7.89	300	0.2712	8.8832
0.0011	10.53	400	0.2693	8.8832
0.0002	13.16	500	0.2760	8.6294
0.0002	15.79	600	0.2853	7.8680
0.0001	18.42	700	0.2866	7.6142
0.0001	21.05	800	0.2889	7.6142
0.0001	23.68	900	0.2904	7.6142
0.0001	26.32	1000	0.2909	7.6142

Framework versions

• Transformers 4.40.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2