pseudolabeling-step2-F04

This model is a fine-tuned version of yip-i/wav2vec2-pretrain-demo on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.2502
• 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: 0.0001
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
74.4163	3.36	500	3.6878	1.0
3.3612	6.71	1000	3.5619	1.0
3.3127	10.07	1500	3.5773	1.0
3.2104	13.42	2000	3.5299	1.0
3.2067	16.78	2500	3.5704	0.9922
3.1511	20.13	3000	4.3842	1.0
3.0825	23.49	3500	4.2644	1.0
3.0959	26.85	4000	5.2502	1.0

Framework versions

• Transformers 4.23.1
• Pytorch 1.12.1+cu113
• Datasets 1.18.3
• Tokenizers 0.13.2