wav2vec2-large-xls-r-300m-tsovatush-demo-colab-bryn-hauk-8-5

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: 3.0703
• Wer: None

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
5.9513	2.7682	400	3.4677	None
3.4754	5.5363	800	3.4689	None
3.316	8.3045	1200	3.2686	None
2.8937	11.0727	1600	2.7787	None
2.441	13.8408	2000	2.7101	None
2.1634	16.6090	2400	2.6361	None
1.9279	19.3772	2800	2.7148	0.9165
1.7059	22.1453	3200	2.5831	None
1.4872	24.9135	3600	2.7741	None
1.2666	27.6817	4000	3.0703	None

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 1.18.3
• Tokenizers 0.19.1