mms-300m-ngen

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

• Loss: 14.2682
• Wer: 1.0

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
13.7013	1.0	27	15.0854	1.0079
13.9308	2.0	54	14.9572	1.0079
14.1021	3.0	81	14.8351	1.0040
14.2026	4.0	108	14.7164	1.0
12.9285	5.0	135	14.6028	1.0
13.6789	6.0	162	14.4974	1.0
13.6224	7.0	189	14.4040	1.0
13.0881	8.0	216	14.3308	1.0
12.838	9.0	243	14.2844	1.0
13.7213	10.0	270	14.2682	1.0

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0
• Datasets 3.6.0
• Tokenizers 0.21.0