MCV_Fleurs_Combined_Irish_ASR_No_Aug

This model is a fine-tuned version of openai/whisper-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2807
• Wer: 24.7066

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• 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: linear
• num_epochs: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4077	0.9174	200	0.5064	46.4497
0.6914	1.8349	400	0.3684	35.7035
0.2925	2.7523	600	0.3187	34.3546
0.1139	3.6697	800	0.3047	28.6330
0.0442	4.5872	1000	0.2987	26.6824
0.0172	5.5046	1200	0.2839	24.7618
0.0077	6.4220	1400	0.2819	26.1809
0.0041	7.3394	1600	0.2806	24.6816
0.0038	8.0	1744	0.2807	24.7066

Framework versions

• Transformers 5.6.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2