whisper-tiny-en_v9

This model is a fine-tuned version of openai/whisper-tiny.en on the common_voice_1_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.6704
• Wer Ortho: 29.3851
• Wer: 20.7154

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-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 100
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.2787	0.1647	100	0.6732	29.2789	20.7342
0.2382	0.3295	200	0.6746	29.2306	20.6731
0.2696	0.4942	300	0.6757	29.1196	20.5793
0.254	0.6590	400	0.6745	29.3223	20.6638
0.2935	0.8237	500	0.6737	29.3658	20.6591
0.2757	0.9885	600	0.6704	29.3851	20.7154

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 2.14.6
• Tokenizers 0.21.1