xls-r-1b-bemgen-balanced-model

This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on the BEMGEN - NA dataset. It achieves the following results on the evaluation set:

• Loss: 0.2854
• Wer: 0.4379

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• 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

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0.3571	100	3.7395	1.0075
No log	0.7143	200	2.8366	1.0
No log	1.0714	300	2.3453	1.0
No log	1.4286	400	0.5266	0.7476
5.5915	1.7857	500	0.4681	0.6964
5.5915	2.1429	600	0.4157	0.6484
5.5915	2.5	700	0.3743	0.5750
5.5915	2.8571	800	0.3286	0.5147
5.5915	3.2143	900	0.3155	0.4767
0.5861	3.5714	1000	0.3057	0.4911
0.5861	3.9286	1100	0.3009	0.4612
0.5861	4.2857	1200	0.2854	0.4382
0.5861	4.6429	1300	0.3016	0.4517
0.5861	5.0	1400	0.2924	0.4434
0.3557	5.3571	1500	0.3079	0.4488
0.3557	5.7143	1600	0.3054	0.4405

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0