wav2vec2-base-librispeech

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

• Loss: 0.2162
• Wer: 0.1419

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: 64
• eval_batch_size: 32
• 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.606	4.13	500	2.0411	0.7943
0.3862	8.26	1000	0.3058	0.2202
0.1253	12.4	1500	0.2450	0.1908
0.0794	16.53	2000	0.2152	0.1531
0.0566	20.66	2500	0.2012	0.1457
0.0446	24.79	3000	0.2061	0.1432
0.0363	28.93	3500	0.2162	0.1419

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3