Model_S_P_Wav2Vec2_Versi2

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.2422
• Wer: 0.6012
• Cer: 0.2488

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

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
3.8621	4.17	400	2.0001	0.8875	0.3540
0.7188	8.33	800	1.7482	0.6799	0.2826
0.4024	12.5	1200	2.0649	0.6533	0.2723
0.2796	16.67	1600	1.9127	0.6452	0.2666
0.2053	20.83	2000	1.8885	0.6095	0.2555
0.1523	25.0	2400	2.0067	0.6168	0.2569
0.105	29.17	2800	2.2422	0.6012	0.2488

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 1.18.3
• Tokenizers 0.13.3