speecht5_finetuned_voxpopuli_nl

This model was trained from scratch on the voxpopuli dataset. It achieves the following results on the evaluation set:

• Loss: 0.4590

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: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• 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
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6863	0.8607	200	0.6124
0.5721	1.7230	400	0.5167
0.5396	2.5853	600	0.4984
0.5289	3.4476	800	0.4868
0.5172	4.3098	1000	0.4815
0.5169	5.1721	1200	0.4771
0.5108	6.0344	1400	0.4740
0.5086	6.8951	1600	0.4715
0.5042	7.7574	1800	0.4699
0.4939	8.6197	2000	0.4678
0.4965	9.4820	2200	0.4667
0.5004	10.3443	2400	0.4644
0.4906	11.2066	2600	0.4617
0.4889	12.0689	2800	0.4612
0.493	12.9295	3000	0.4601
0.4893	13.7918	3200	0.4599
0.4894	14.6541	3400	0.4600
0.4922	15.5164	3600	0.4594
0.491	16.3787	3800	0.4599
0.482	17.2410	4000	0.4590

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1