Naji101/whisper-tiny-en_v8

Whisper tiny En v6 Naji

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 1 dataset. It achieves the following results on the evaluation set:

• Loss: 1.3391
• Wer Ortho: 65.9668
• Wer: 58.5974

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: 3e-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: 500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
3.3144	0.0824	50	3.4046	31.2868	22.5696
1.8588	0.1647	100	2.2137	32.0108	22.2786
1.4358	0.2471	150	1.8701	30.1236	21.2130
1.2467	0.3295	200	1.6781	33.4154	24.1797
1.1247	0.4119	250	1.5011	48.5520	39.9286
0.9712	0.4942	300	1.3391	65.9668	58.5974

Framework versions

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