baoule

This model is a fine-tuned version of openai/whisper-small on the abdouaziz/baoule2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2863
• Wer: 0.1696

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-05
• train_batch_size: 32
• eval_batch_size: 32
• 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: 48000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.4736	2.4510	500	0.2696	0.3169
0.0515	4.9020	1000	0.2703	0.1995
0.0185	7.3529	1500	0.2761	0.2021
0.0112	9.8039	2000	0.2853	0.2609
0.0083	12.2549	2500	0.2922	0.1829
0.0073	14.7059	3000	0.2884	0.1711
0.0066	17.1569	3500	0.2849	0.1724
0.0063	19.6078	4000	0.2863	0.1696
0.0053	22.0588	4500	0.2889	0.1768
0.0053	24.5098	5000	0.2951	0.1757
0.0047	26.9608	5500	0.2844	0.1837

Framework versions

• Transformers 4.46.0
• Pytorch 2.7.0+cu126
• Datasets 3.3.2
• Tokenizers 0.20.3