wav2vec2-large-mms-1b-uyghur-latin

This model is a fine-tuned version of facebook/mms-1b-all on the None dataset. It achieves the following best results on the evaluation set:

• Best Wer: 30.8949%
• Best Cer: 5.9823 %

Training procedure

Finetuning code avaiblable in https://github.com/ixxan/ug-speech

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Cer Ortho
0.3425	1.0006	1313	0.3081	35.3122	6.8424
0.3218	2.0011	2626	0.2771	31.7204	6.1840
0.3012	3.0017	3939	0.2739	30.8949	5.9823
0.2961	3.9989	5248	0.2771	31.7116	6.1806

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3