XLRS_MediumDataset

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3156
  • Wer: 0.2796

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1000
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Wer
4.6662 1.54 500 3.1089 1.0
2.6856 3.08 1000 1.1527 1.0257
0.8295 4.62 1500 0.3830 0.4426
0.4529 6.15 2000 0.3183 0.3827
0.3135 7.69 2500 0.2764 0.3517
0.2652 9.23 3000 0.2630 0.3145
0.2065 10.77 3500 0.2679 0.3141
0.1802 12.31 4000 0.3010 0.3057
0.1557 13.85 4500 0.2971 0.3017
0.1438 15.38 5000 0.2953 0.3020
0.1206 16.92 5500 0.3185 0.2959
0.1104 18.46 6000 0.3114 0.2871
0.1005 20.0 6500 0.3230 0.2873
0.092 21.54 7000 0.3130 0.2844
0.0946 23.08 7500 0.3130 0.2837
0.09 24.62 8000 0.3189 0.2787
0.0822 26.15 8500 0.3200 0.2783
0.0805 27.69 9000 0.3247 0.2777
0.0827 29.23 9500 0.3156 0.2796

Framework versions

  • Transformers 4.17.0
  • Pytorch 2.5.1+cu121
  • Datasets 1.18.3
  • Tokenizers 0.20.3
Downloads last month
3
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.