whisper-medium-v3-ff4

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

• Loss: 0.2572
• Wer: 0.1694

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 26000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3543	0.6393	5000	0.3380	0.2361
0.2259	1.2786	10000	0.2893	0.2022
0.1938	1.9179	15000	0.2597	0.1790
0.1119	2.5572	20000	0.2572	0.1694
0.0632	3.1965	25000	0.2615	0.1660

Framework versions

• Transformers 4.46.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.20.3