wav2vec2-xlsr-53-espeak-cv-ft-evn2-ntsema-colab

This model is a fine-tuned version of facebook/wav2vec2-xlsr-53-espeak-cv-ft on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: 2.0299
• Wer: 0.9867

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• 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
4.2753	6.15	400	1.6106	0.99
0.8472	12.3	800	1.6731	0.99
0.4462	18.46	1200	1.8516	0.99
0.2556	24.61	1600	2.0299	0.9867

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1+cu113
• Datasets 2.6.1
• Tokenizers 0.13.2