mms-300m-ach-kadima

This model is a fine-tuned version of facebook/mms-300m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8482
• Wer: 0.3784
• Cer: 0.1490

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
21.4759	4.5473	500	2.7033	1.0	1.0
4.3679	9.0912	1000	0.6118	0.4074	0.1605
2.9515	13.6385	1500	0.6764	0.4073	0.1714
2.0563	18.1824	2000	0.7575	0.3699	0.1494
1.4838	22.7298	2500	0.8482	0.3784	0.1490

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2