mms-300m-mas-gbotemi

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: 0.5609
• Wer: 0.5135
• Cer: 0.1127

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
23.1415	2.5126	500	2.8730	1.0	1.0
22.6111	5.0251	1000	2.8362	1.0	1.0
5.5397	7.5377	1500	0.6171	0.6447	0.1574
3.2974	10.0503	2000	0.4744	0.5189	0.1166
2.5712	12.5628	2500	0.4465	0.4989	0.1110
2.0020	15.0754	3000	0.4694	0.5019	0.1112
1.7115	17.5879	3500	0.5107	0.5100	0.1124
1.3523	20.1005	4000	0.5609	0.5135	0.1127

Framework versions

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