whisper-tiny-en

This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.9035
• Wer Ortho: 0.354643
• Wer: 0.357235

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.0005	35.71	500	0.7515	36.1373	36.4341
0.0002	71.43	1000	0.8095	36.4065	36.5633
0.0001	107.14	1500	0.8421	36.4738	36.6925
0.0001	142.86	2000	0.8636	35.4643	35.5943
0.0001	178.57	2500	0.8822	35.6662	35.7235
0.0	214.29	3000	0.8931	35.4643	35.7235
0.0	250.0	3500	0.9013	35.4643	35.7235
0.0	285.71	4000	0.9035	35.4643	35.7235

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3