whisper-small-lm

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

• Loss: 0.0856
• Wer: 0.0857

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• 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: 500
• num_epochs: 30.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2062	1.7342	500	0.1897	0.1699
0.056	3.4710	1000	0.0987	0.1042
0.0207	5.2078	1500	0.0887	0.0915
0.0161	6.9420	2000	0.0856	0.0857
0.007	8.6788	2500	0.0896	0.0842
0.0055	10.4156	3000	0.0908	0.0840

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1