wav2vec2-xlsr-53-espeak-cv-ft-evn6-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: 1.2335
• Wer: 0.9431

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
3.847	4.0	400	0.9836	0.9933
0.8626	8.0	800	0.8241	0.9666
0.536	12.0	1200	0.9166	0.9565
0.3374	16.0	1600	1.1043	0.9732
0.2251	20.0	2000	1.1423	0.9632
0.1649	24.0	2400	1.1648	0.9599
0.1244	28.0	2800	1.2335	0.9431

Framework versions

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