w2v2-libri-10min

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: 2.0535
• Wer: 0.5781

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 2500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
4.9227	62.5	250	2.9488	1.0
1.0088	125.0	500	1.5970	0.6570
0.0694	187.5	750	1.6500	0.6680
0.0585	250.0	1000	1.8617	0.6501
0.0226	312.5	1250	2.3323	0.6362
0.0175	375.0	1500	2.0121	0.6113
0.0093	437.5	1750	2.0362	0.5781
0.006	500.0	2000	2.1563	0.5989
0.0042	562.5	2250	2.1231	0.5837
0.0027	625.0	2500	2.0535	0.5781

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 1.18.3
• Tokenizers 0.13.3