audiox_mucs_mucs_49234_trial

This model is a fine-tuned version of jiviai/audioX-north-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2711
• Global Wer: 438.1538

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: 5e-06
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.4787	0.0134	50	0.4361	29.0038
0.2924	0.0268	100	0.3856	31.8066
0.2740	0.0402	150	0.3279	35.5032
0.2460	0.0536	200	0.2992	86.0973
0.2224	0.0670	250	0.3067	28.9936
0.2170	0.0804	300	0.3010	71.8798
0.2364	0.0938	350	0.2831	71.4532
0.2511	0.1072	400	0.2653	194.5567
0.2646	0.1206	450	0.2674	60.4245
0.2223	0.1340	500	0.2711	438.1538

Framework versions

• Transformers 5.0.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 3.6.0
• Tokenizers 0.22.2