wav2vec2-large-xls-r-300m-dysarthria

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0615
• Wer: 0.1764

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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
16.998	2.17	400	3.4205	1.0
3.6507	4.34	800	3.2819	1.0
3.2148	6.5	1200	3.0239	1.0
2.8464	8.67	1600	2.5810	1.0
2.3923	10.84	2000	2.2368	1.0
1.9358	13.01	2400	1.7072	1.0
1.5043	15.18	2800	1.3435	1.0
1.1169	17.34	3200	0.8979	0.9701
0.749	19.51	3600	0.5764	0.7490
0.4855	21.68	4000	0.2876	0.4763
0.2902	23.85	4400	0.1645	0.3379
0.198	26.02	4800	0.0988	0.2307
0.1358	28.18	5200	0.0615	0.1764

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1